Revenge of the 2001 Dot-Com Nerds. They are making the case for 25-year DCFs and terminal values still being a thing. $NOK $CIEN $CSCO $BB https://t.co/b6m5M1iTyf
Let’s be real here… Every AI data center Nvidia ever builds needs optical transceivers to move data between GPUs. Not because of GPU performance. Not because of CUDA software. Not because of rack density. Because of physics. Electrical signals cannot move data at the bandwidth AI clusters require over distances greater than a few meters. Light can. $INTC own data center documentation confirms fiber optic interconnects at 800G+ are required for every GPU cluster at scale. Demand for 800G transceivers is STRONG with volume ramps underway in 2026. $AAOI is still ONLY approximately $15B compared to peers as $LITE $COHR $CIEN at +50$B. The GPU market is $500B+. The company supplying the optical layer those GPUs depend on for data movement is at $15B. Rosenblatt raised their price target to $220 in May 2026, up from $140 in March. Current price implies limited upside on Rosenblatt’s numbers relative to earlier valuations. Now execution matters.Either the AI GPU capex cycle delivers the optical component volume the data suggests, or it doesn’t. At current market cap vs. the $500B+ infrastructure build, valuations reflect much of the outcome. But when execution follows $AAOI will easily run past $280, $300, $320, $350, $400. I’m long $AAOI. -BP Save this. Not financial advice. DYOR.
$NOK Unusual options activity keeps pouring into $NOK and it’s getting harder for the market to ignore what is happening beneath the surface. Over $10M in LEAPS calls have hit Nokia in the last 2 weeks alone. The biggest trade? 50,000 contracts on the Dec 2026 $24 calls worth roughly $4.15M. Someone just made a multi-million dollar bet that Nokia trades 66% higher within 7 months. This is not the old Nokia the market remembers. The story changed after the Infinera acquisition. Nokia now controls one of the most vertically integrated optical networking stacks in the industry: → Photonic chips (PICs) → Optical components → Complete networking systems Most competitors only own one layer of the stack. Nokia is now positioned directly in the center of the AI infrastructure buildout, optical networking expansion, AI-RAN deployments, hyperscaler traffic growth, and future 6G infrastructure. Meanwhile, the legacy mobile business is still suppressing the valuation multiple even as optical and AI networking become a larger percentage of revenue every year. That disconnect is exactly why the options market may be getting aggressive here. The market is rewarding optical/photonics companies with massive forward multiples: $AXTI ~295x $CIEN ~112x $AAOI ~100x $COHR ~72x $LITE ~60x Nokia still trades at a major discount despite arguably having one of the strongest strategic positions in optical infrastructure. The options chain is also revealing a very important setup: $14 = support floor $15 = trigger level Dealers are heavily positioned around those strikes. If $NOK decisively breaks above $15 with volume, gamma dynamics can quickly shift from resistance into forced dealer buying. That is why June $15 calls now carry the largest open interest on the chain with more than 120K contracts open. At the same time, traders continue stacking farther-out calls: → Jul $20C → Sep $21C → Dec $23C → Jan 2027 $20C This looks less like random speculation and more like institutions positioning for a full re-rating cycle tied to AI networking, optical infrastructure, and AI-RAN adoption. Still volatile and still speculative, but definitely one of the more interesting asymmetric setups in AI infrastructure right now. https://t.co/ErU6FSMeNt
$NOK multi-decade high. Still a long way to peak internet days (much like $CIEN ), but it is working its way there. https://t.co/dTQ2Yttmyq
$CIEN banging out a new all-time today. Up +8%. This is one that I have a high degree of confidence will continue to trade higher from here. Earnings on 6/5/26. https://t.co/cF65NSiCLU
The Complete Data Center Stack: Where the AI Infrastructure Money Flows Most investors still think AI is just about GPUs. That’s incomplete. AI is an infrastructure buildout, and the real opportunity spans the entire data center stack. Every inference, every training run, and every deployed model depends on multiple layers working together. Here’s the breakdown: 1. Compute Silicon (The Brain) Tickers: $NVDA, $AMD, $AVGO, $INTC This is the foundation. GPUs, CPUs, accelerators, and custom silicon power training and inference. Why it matters: - Compute demand keeps rising with larger models - AI workloads are forcing faster chip innovation - Custom ASICs are becoming a major trend 2. Server OEMs & Solutions (The Hardware Layer) Tickers: $SMCI, $DELL, $HPE, $VRT, $ETN, $MOD Chips need systems. These companies assemble and deliver the physical AI servers and power systems. Why it matters: - AI racks are denser and hotter - Power distribution is now critical - Cooling is becoming a competitive advantage 3. Memory & Storage (The Hidden Bottleneck) Tickers: $SNDK, SK Hynix, $MU, $WDC, $P, Samsung, $NTAP AI models consume massive amounts of memory bandwidth and storage. Why it matters: - High-bandwidth memory is becoming strategic infrastructure - Data storage demand rises with AI deployment - Faster access = better model performance 4. Networking & Connectivity (The Nervous System) Tickers: $ANET, $CSCO, $MRVL, $CRDO, $CIEN, $NOK AI clusters must communicate at ultra-high speed. Why it matters: - Faster networking reduces latency - Data movement is becoming expensive - Scale depends on interconnect efficiency Key idea: AI cannot scale without bandwidth. 5. Neoclouds & Physical Infrastructure (The New Builders) Tickers: $NBIS, $IREN, $CRWV, $APLD $CIFR $DGXX These companies provide specialized AI infrastructure and hosting. Why it matters: - Cloud alternatives are growing - AI-native infrastructure is becoming valuable - Capacity shortages create pricing power 6. Energy (The Ultimate Constraint) Tickers: $CEG, $NEE, $EOSE, $GEV, $EQT, $VST $OKLO $BE $FLNC AI consumes enormous electricity. Power availability is becoming a limiting factor. Why it matters: - Grid demand is surging - Battery storage is essential - Reliable baseload power matters Final Thought The market often focuses on one winner. But AI infrastructure is an ecosystem. If you want to understand where capital flows next, follow the stack: Compute → Servers → Memory → Networking → Infrastructure → Energy The biggest winners in the AI cycle may not always be the obvious names. Sometimes the best opportunities are in the supporting layers that make the whole system possible.
FIBER / BANDWIDTH: The Nervous System of AI AI isn’t just built on chips. It moves on bandwidth. Every model update, inference query, and real-time AI decision travels through fiber. Without high-speed connectivity, compute is useless. Why it matters: → More AI agents = more data traffic → More inference = higher bandwidth demand → Fiber = the last mile of intelligence Watch: $FYBR · $LUMN · $CSCO · $INFN · $CIEN · $ASTS The AI boom isn’t just compute. It’s connectivity.
The Complete Semiconductor Playbook — AI Supercycle 2026 The AI trade is no longer one ticker. It’s a full stack. Here’s every layer: 💾 MEMORY $MU · $WDC · $STX · $SNDK · $AMAT · $LRCX → HBM4 = the new oil. Not a commodity anymore. Strategic infrastructure. 🔌 CONTROLLERS $RMBS · $SIMO · $ATOM · $MCHP · $ADI · $TXN → The traffic cops of every AI system. Overlooked. Undervalued. Critical. 📦 PACKAGING / TESTING $AMKR · $FORM · $CAMT · $ONTO · $COHU · $ACMR → CoWoS. HBM stacking. Chiplets. This is where chips become systems. 🧠 NEXT-GEN MEMORY $MRAM · $GSIT · $NVEC · $NLST · $CEVA → Beyond DRAM. Early-stage. Asymmetric upside when the cycle arrives. 🌐 OPTICAL / INTERCONNECT $LITE · $AAOI · $COHR · $CIEN · $MRVL · $GLW · $MTSI · $TSEM · $POET · $LWLG · $FN → The hottest sub-sector in semis right now. 800G → 1.6T. CPO is the next arms race.
$CIEN someone got the memo. All-time high today. and it keeps on going... https://t.co/dxMuqFWXoe
$LITE In Lasers We Trust! They crushed it. Revenue growth, margin expansion, and strong guidance. Strength in scale-across a potential positive read through for $CIEN $NOK. --------------- “Lumentum delivered an exceptional third quarter, with revenue growing 90% year over year to a record $808 million. While our top line growth continues to garner headlines, the more impressive part of our recent performance has been our margin expansion. In íscal Q3, gross margin improved by 540 basis points on quarter and operating margin by 700 basis points,” said President and CEO Michael Hurlston. “Margin improvement has been driven by many factors including operational rigor, pricing discipline, and product mix. This quarter, mix was aided by strength in laser chips, but also by a less-heralded part of our portfolio, ‘scale-across’ components, which include our pump lasers and narrow linewidth laser assemblies. As our key growth drivers of co-packaged optics and optical circuit switches begin to kick in, we would expect further increases in earnings power.” Call in 10 mins. Let's see what more they have to say...
The Full Optical AI Industry Chain: 1. IC Design & Manufacturing (Where the brains are built) ASIC / xPU / Photonic ICs → $NVDA $MRVL $AVGO $LITE $COHR $INTC $AMD $CSCO Wafer Foundries → $TSM $TSEM $GFS $UMC 2. Materials (The Overlooked Layer) (The foundation of photonics) Indium Phosphide → $AXTI $IQE Gallium Arsenide → $AXTI $MTSI This layer is underfollowed and could see strong upside as optical demand expands. 3. Optical Components (The data highways) VCSEL / EML Lasers → $LITE $COHR $AVGO $AAOI Optical Module Assembly → $FN $COHR $LITE $MRVL $CSCO Fiber / WDM / Optical Connectivity → $CIEN $GLW $LITE MPO Connectors → $GLW $APH Optical Fiber → $GLW 4. Packaging & Testing (The hidden backbone) Packaging → $ASX $AMKR $FN $JBL Packaging Equipment → $KLIC Testing → $TER $KEYS $FORM $AEHR
Optical Communications: The Next Breakout in AI Infrastructure Everyone is focused on GPUs, but the real bottleneck in AI is no longer compute — it’s data movement. As AI clusters scale into hundreds of thousands of GPUs, the challenge becomes moving data faster, farther, and with less power. Copper is hitting its limits. Optical is the next layer of scale. Why optical matters: • Higher bandwidth • Lower latency • Lower power consumption • Less heat • Better long-distance transmission That’s why hyperscalers are aggressively investing in optical-heavy infrastructure. Core optical AI plays on my radar: $COHR — broad optical exposure $LITE — datacenter optical leader $AAOI — optical modules growth $FN — manufacturing backbone $MRVL — AI networking silicon $CIEN — optical transport infrastructure AI compute gets the attention. Optical networking moves the data. Without optical interconnects, AI doesn’t scale. Follow the data pipes, not just the chips. Not financial advice.
Where NOK Sits vs. The Photonics Stack- Most photonics plays are component-level: $COHR, $LITE, $CIEN, $AAOI, $MRVL, $FN, $POET, $LWLG $NOK is the system-level integrator sitting above that entire stack — buying from it, assembling it, and deploying it at hyperscaler + telco scale globally. It’s the platform layer of the optical AI infrastructure trade.
$CIEN is another one. Multi-decade high today. Now $75bn EV from $60bn EV when I posted this on 3/24/26. Still feels cheap at $75bn EV. I'm long 1/21/28 calls. Yes, there is competition, and it is still very difficult to replicate what they do at their scale plus quality. Earnings on 6/5/26, so we'll need to wait until then to find out more.
Photonics & Semiconductors: The Quiet Kings of This Market The rotation is happening in plain sight. While macro noise dominates headlines, photonics and semiconductor names are quietly building the infrastructure that every AI workload, data center, and next-gen network depends on — and the market is starting to reprice that reality. $COHR, $LITE, $MRVL, $CIEN, $INTC, $SNDK , $AXTI, $MU , $AAOI, and $SIMO aren’t just riding a trend; they’re supplying the rails that the entire AI buildout runs on. Optical interconnects, silicon photonics, and coherent transceivers aren’t optional upgrades — they’re the physical bottleneck every hyperscaler is scrambling to solve right now. On the semiconductor side, the thesis is just as structural: demand for specialized chips across AI inference, optical engines, and edge compute isn’t slowing; it’s accelerating into 2026 and beyond. When bandwidth demand grows exponentially and copper simply can’t keep up, light wins — and the companies enabling that shift are sitting at the intersection of two of the most durable capex cycles in market history. This isn’t hype. This is infrastructure.
$CIEN all time high up ~4% this morning. https://t.co/aynXzbt2Dw
$CIEN — The Optical Plumbing That Moves Data Between Data Centers Ciena plays the transceiver and systems layer, focused on high-capacity routing and network equipment. As AI traffic between hyperscaler campuses explodes, Ciena’s coherent optical networking gear becomes the highway connecting them. The inter-datacenter layer is as critical as inside-rack optics.
My Photonics watchlist:- Materials/Substrates: $AXTI $GLW $IQEPF Foundry/Manufacturing: $TSEM $FN $GFS $MKSI Components/Lasers: $COHR $LITE $AAOI $SIVE Silicon/DSP: $AVGO $MRVL $MTSI $CRDO Systems/Networks: $CIEN $NOK $CLS Infrastructure: $NVDA Moonshots: $POET $LWLG
$CIEN is ascending to an all-time high this morning. Looks like it is picking up steam again after taking a breather. https://t.co/qL7AVHuJBN
$AAOI $LITE $COHR $CIEN @CitronResearch next time you want to short a GAI infrastructure name, message me first. I might save you some time, heartache, and money. https://t.co/2RewZsHQ3j
Most investors are always chasing the last wave. The AI infrastructure cycle is playing out in layers — and each one is bigger than the last. Here’s the full roadmap: Wave 1: Semiconductors ✅ (Priced In) $NVDA. $AMD. $AVGO. The GPU arms race. Everyone knows this trade. Raw compute was the foundation. Most of the upside? Already captured. Wave 2: Memory & Storage ✅ (Confirming) More data. More inference. More throughput. Flash, DRAM, HBM all followed. $SNDK ’s recent move is the latest confirmation. The storage bottleneck is real — and the market is finally catching up. Wave 3: Photonics & Optical Networking 👀 (In Motion NOW) This is what most retail investors are still sleeping on. You cannot move AI workloads at scale with copper wire. Data centers need ultra-fast, low-latency interconnects — and photonics is the only answer. Silicon photonics is actively replacing copper-based communication inside and between hyperscaler campuses. The plays institutions are quietly accumulating: $AAOI — pure-play optical components, deep data center exposure $COHR — vertically integrated photonics at scale $LITE — transceivers powering the hyperscaler build-out $POET — integrated optical engines, early-stage, asymmetric upside $CIEN — backbone optical networking infrastructure $FN — precision optical manufacturing, best-in-class margins This wave is not coming. It is here. Wave 4: Power & Energy Infrastructure (Early Stage) AI doesn’t run on ambition. It runs on electricity. Data centers are projected to consume 8–10% of U.S. power by 2030. That is not a tailwind. That is a structural demand shock the grid is not prepared for. Nuclear is the only baseload solution that can sit next to a hyperscaler campus and deliver consistent, carbon-light power at scale. The plays: $CEG & $VST — utilities with nuclear-powered AI offtake contracts $CCJ — uranium supply, the fuel behind the revival $OKLO & $SMR — next-gen small modular reactors $VRT — liquid cooling and power management inside data centers The capex commitments from $MSFT, $AMZN , and $GOOG are already locked in. The power infrastructure just needs to catch up. This wave is early. That is the opportunity. Wave 5: Robotics 🤖 (Loading…) Once AI is trained, powered, and connected — it needs a body. Physical deployment into warehouses, factories, logistics, and healthcare. This is the final and potentially the largest layer of the entire cycle. Hardware-software integration is still 12–24 months from full ignition. But the smart money is already laying the groundwork quietly. The pattern is simple: Each layer enables the one after it. → Semis built the brain → Memory gave it recall → Photonics gave it nerves → Power gives it fuel → Robotics gives it hands The investors winning this decade are not chasing yesterday’s leader. They are identifying tomorrow’s bottleneck — and positioning before the crowd arrives. The question is not which wave already ran. The question is: which wave are YOU in front of? $AAOI $COHR $LITE $POET $CIEN $FN $CEG $VST $OKLO $SMR $CCJ $VRT DOYR , bookmark this and retweet for others.
PHOTONICS / OPTICAL STACK — THE FULL VALUE CHAIN “The layer below the chips” — AI’s real bottleneck Layer 1 — Materials Foundation of everything (InP, fiber, wafers) $AXTI $GLW $COHR Layer 2 — SiPh Foundries Where photonic chips are fabricated $TSMC $GFS $TSEM Layer 3 — Lasers (Bottleneck) EML + CW lasers — tightest supply in the chain $LITE $COHR $AAOI $FN Layer 4 — PICs (Optical Chips) Integration layer replacing discrete components $POET $LWLG $MRVL $AVGO Layer 5 — Transceivers 400G → 800G → 1.6T data movement $CIEN $AAOI $LITE $COHR Layer 6 — Co-Packaged Optics (CPO) Next-gen architecture — optics + silicon combined $NVDA $AVGO $MRVL $COHR $LITE $ALAB Layer 7 — Optical Switching (OCS) Routing light without electrical conversion $CIEN $COHR $LITE Layer 8 — Networking Systems DCI, long-haul, metro infrastructure $CIEN $ANET $NOK Layer 9 — Test & EDA Hidden backbone enabling scale $KEYS $FORM $ANSS Layer 10 — Demand Drivers Hyperscalers pulling the entire chain $NVDA $META $GOOGL $MSFT $AMZN THE BIG PICTURE (2026–2027) - AI demand is exponential - Electrical interconnects are hitting limits - Optical = the only scalable solution Key Constraint: Laser supply (InP → EML/CW) → Bottleneck today = pricing power tomorrow This isn’t a single stock trade. It’s a full-stack infrastructure shift — like semis in the early 2000s. The winners won’t just be chipmakers… They’ll be the ones enabling light-speed data movement. Not financial advice. DYOR. Bookmark this, retweet for others
PHOTONICS — THE MOMENTUM TRADE SMART MONEY IS ROTATING INTO Institutional capital is quietly flowing into optical networking. This isn’t hype — it’s AI infrastructure. Why photonics is leading right now → AI data centers are hitting a copper bandwidth wall → Optical interconnects = faster, scalable, lower power (light > electrons) → Next-gen systems are being built around this shift → 800G+ demand is accelerating faster than supply can match The supply-demand imbalance → Capacity is tight across key components → Long lead times = sustained demand visibility → Pricing power shifts to suppliers → This is how multi-month sector runs start The names to watch → $LITE — core laser tech, supply constrained, strong demand tailwind → $COHR — scale player, diversified + aligned with AI ecosystem → $FN — manufacturing backbone, major beneficiary of volume ramp → $CIEN — network infrastructure layer, hyperscaler exposure → $MRVL — silicon photonics + custom AI data center chips → $AEHR — testing demand ramp tied to new photonics cycle $POET , $LWLG , $SIVEF , $AAOI , $AXTI What confirms the trend → Breakouts across multiple names — not just one → Moves happening without headlines = institutional accumulation → Strong relative performance vs $SPY Don’t overcomplicate it → This is a sector move, not a single-stock story → You don’t need perfect entries — you need the right theme → Momentum phases last longer than most expect The play → Focus on strength within the sector → Let momentum + capital flows work → Stay until relative strength breaks — then rotate Photonics is the plumbing of AI. And this cycle is just getting started. Not financial advice. DYOR.
Nvidia invested $6B into photonics in March alone. The AI bottleneck isn't chips. It's moving data between them. 10 photonics stocks I'm watching: $AAOI - Applied Optoelectronics $LITE - Lumentum $COHR - Coherent $FN - Fabrinet $MTSI - MACOM Technology $CRDO - Credo Technology $LASR - nLight $AXTI - AXT Inc $MRVL - Marvell Technology $CIEN - Ciena NFA DYOR
$LITE $COHR $CIEN $AAOI As I said two days ago, don't short GAI optical networking names. The momentum is too strong, and one hyperscaler deal announcement will kill you. The stock will run away, and you will be forced to cover at the worst possible time. If you think the price action is BS (which I generally don't), simply sit on the sidelines. If you do short it, I will probably happily take the other side of the trade. $LITE $COHR $CIEN $AAOI
$LITE $COHR $CIEN $AAOI (Bloomberg) -- Lumentum Holdings Inc. said demand from the biggest US tech companies for its optical components is accelerating and on track to fill its order books through 2028. “The capex numbers from the US hyperscalers are enormous and there seems to be no end in sight,” Chief Executive Officer Michael Hurlston said in an interview in Tokyo Friday. “We’re falling further and further behind the demand. We would be sold out though all of 2028 within two quarters.” Hurlston’s projection extends the company’s previous disclosure that it’s sold out until at least the end of 2027. It underscores expectations that demand for equipment to fill artificial intelligence data centers will remain resilient, even as the war in Iran rattles oil markets and the global economy. The company, which last month received an investment of $2 billion from Nvidia Corp., alongside rival Coherent Corp., supplies advanced indium phosphide devices that enable high-speed data transmission. Its Nasdaq-traded shares have risen more than 1,500% over the past year, as optoelectronics are increasingly showing up in cloud computing clusters. To meet runaway demand, San Jose, California-based Lumentum has ramped up capacity 12-fold at its key factory in the Tokyo metropolitan area over the past two years. It now plans to invest at least $100 million in the site and a neighboring facility. “We could see it probably going up to even a quarter billion,” the CEO said. “You cannot imagine the pressure that’s on the team to ramp up. All the US hyperscalers want to know how many units are we going to be able to make this quarter, next quarter, the quarter after; Because it’s really the bottleneck right now for the industry.” Silicon photonics, which uses light to transmit data, is expected to replace copper-based interconnects in data center clusters as advanced AI hardware requires ever faster transmission. Optical networking also helps systems generate less heat and consume less power. Japan is home to companies with expertise in indium phosphide, fiber optic materials and silicon photonics, including such companies as Fujikura Ltd., Sumitomo Electric Industries Ltd. and Hamamatsu Photonics KK. Lumentum acquired some of its technology via Japan through its 2018 purchase of Oclaro Inc. which had inherited Hitachi Ltd.’s optical research. Today, its Sagamihara facility is one of the world’s most advanced indium phosphide device production sites, capable of delivering stable, high-quality output. Room to add more production equipment at the Sagamihara factory and the neighboring Takao site is limited, and Lumentum is scouting new locations in Japan to expand output. It’s targeting older electronics fabrication facilities that it can repurpose — which would shorten the time needed to bring new capacity online. The company considered acquiring one such site in Japan last year, Hurlston said. Moving production out of Japan is not feasible as indium phosphide manufacturing requires precise control of variables such as temperature, humidity, mixing speeds and water quality, the CEO said. A rich web of suppliers surrounding the material is also key, and the company has signed a seven-year arrangement with one Japanese supplier that Hurlston described as “very critical.” “There’s a real strength in Japan around photonics, around these optoelectronics,” Hurlston said. “It’s been there for 40 years. It developed here. The technology estate here. So there’s a whole ecosystem of suppliers that go around that.” Despite all the investment in factories, long-term supplier contracts and hiring more engineers, Lumentum expects its backlog of orders to keep piling up. “There’s no way this is going to continue forever, it’s impossible, but there seems to be a decent amount of sustainability in the cycle for at least five years,” Hurlston said. “When we talk about sold out, that means non-cancelable agreements. It’s a big thing.”
GAI Optical networking is still going nuts to start the day. $LITE $COHR $CIEN $AAOI https://t.co/ilJLQmA0Nj
Photonics / Optical / Semiconductor — Full Stack View Foundries $TSEM $GFS → Scaling silicon photonics capacity to meet demand surge Optical Components & Lasers $COHR $LITE $IPGP $LASR → Core building blocks: lasers, modulators, photodiodes Transceivers & Optical Engines $AAOI $POET → Where electrical meets optical — key bottleneck + opportunity Chip Design & SiPh Integration $MRVL $MTSI → Designing the brains behind optical interconnects Manufacturing & Packaging $FN → Advanced assembly becomes critical as complexity rises Networking & Systems $CIEN $GLW → Moving massive data through fiber at scale Testing & Materials $AEHR $AXTI $LWLG → Enabling performance, reliability, and next-gen efficiency Key Drivers → AI data center buildout (800G → 1.6T → CPO) → Demand > supply through 2027 → Aggressive M&A across the ecosystem This isn’t just a sector — it’s an entire supply chain being repriced in real time. Photonics isn’t coming… it’s already here. Not financial advice. DYOR.
Optical networking got the internet going nuts. It is all about lasers. $LITE $COHR $CIEN $AAOI https://t.co/TzFytnAE0x
Do yourself a favor and don’t short any part of the GAI optical networking ecosystem. The bullish momentum itself is incredibly strong. One hyperscaler deal announcement and you will get your face ripped off. $LITE $COHR $CIEN $AAOI
@aleabitoreddit Thank you, man. I appreciate it. From what I currently see, I’m long-term holder of $CIEN. I love your feet on the street visit to CS 153. That’s solid diligence. I can’t believe there are empty seats in that classroom; those students don’t understand what they are missing.
@TheValueist Great callout on $CIEN, I didn't long personally but glad those leaps played out for you.
Ripper and 52w high for $CIEN (and many others) in the night session. Let’s hope it holds. https://t.co/c1wOOZMFgK
$CIEN EXECUTIVE OVERVIEW The core factual claim in the post is verified. Ciena disclosed on 2026-03-24 that Lightstorm upgraded the Japan-Guam-Australia, or JGA, submarine cable to support 400Gbps services using WaveLogic coherent optical technology, and that the upgraded route is already carrying production traffic for a leading global cloud platform between Japan and Australia. Ciena also stated that JGA offers an alternative to traditional Asia-Pacific routes transiting the South China Sea, with improved Tokyo-Sydney latency and greater operational and geopolitical diversity. Lightstorm’s own APAC materials separately describe JGA as a sub-97 ms Tokyo-Sydney route, targeted at cloud, AI, gaming, and financial-services workloads, and note a 2024 Ciena technology upgrade on the route. The latest disclosure is therefore best interpreted as commercial confirmation of a live production deployment rather than the first appearance of the underlying capability.  The post is directionally correct on the strategic read-through but over-precise on several economic and valuation claims. The strongest conclusion is that AI and cloud demand are now visibly influencing subsea and long-haul optical upgrade cycles, and that Ciena has real leverage to that spend. The weakest conclusions are the assertion that the industry crossed a discrete 400G economic threshold in late 2025, the generic claim of roughly 40% power reduction, and the implication that this single event establishes a secure, highly visible, margin-expanding runway. What is demonstrably true is that the broader optical market has already turned sharply upward: Dell’Oro estimates the optical transport market grew 10% in 2025, DCI WDM purchases grew nearly 40%, and direct cloud-provider optical purchases grew around 50%.  VALIDATION OF WHAT THE EVENT ACTUALLY MEANS The technical significance is narrower and more important than the post suggests. This was not a new cable build. It was an upgrade in service rate and transmission efficiency on an existing subsea asset. The economic value of moving traffic from 100G to 400G on installed wet plant is that more usable capacity can be extracted from sunk subsea infrastructure, landing-station space, and associated terrestrial interconnect without laying new fiber. Ciena’s own materials on coherent evolution explicitly describe the benefit of repurposing existing fiber plant to carry more capacity and deferring the need for new fiber builds. In practical investment terms, the event validates a recurring upgrade model on installed subsea routes rather than a one-off greenfield construction cycle.  The post’s statistic that undersea cables carry over 95% of global internet traffic is directionally valid but phrased imprecisely. The more standard formulation in policy and industry sources is that subsea cables carry more than 95% of international or intercontinental data traffic. That distinction matters because a large share of domestic data traffic in major countries remains terrestrial. The post therefore overgeneralizes the statistic slightly, but the underlying strategic conclusion remains correct: submarine infrastructure is still the critical physical layer for cross-border cloud, data, and communications traffic.  The 400G economics section of the post is only partially verified. Official sources support the broad idea that 400G is superior to aggregating 4 x 100G in cost-per-bit, port efficiency, density, and power consumption. HPE states that a single 400G port with optics costs less and consumes less power than 4 individual 100G ports with optics. Cisco provides a concrete optics example of 12W for a 400G breakout module versus 17.2W for 4 discrete 100G optics, which is about 29% lower optical power. Ciena has separately shown much larger long-distance savings in specific transport configurations, including 75% more fiber capacity, 87.5% footprint savings, and 80% power savings when deploying 400G wavelengths rather than 100G wavelengths over long-haul links. The post’s roughly 40% power-reduction figure is therefore plausible in some architectures but not defensible as a universal industry constant. More importantly, no high-quality neutral source was identified that defines a discrete “late 2025” industry crossover point. The correct framing is that 400G became economically compelling across a widening set of optical use cases by 2025, not that the industry crossed a single dated threshold.  The claim that 400G immediately creates direct margin expansion for hyperscalers is too strong. The more defensible statement is that higher-rate coherent upgrades reduce network TCO, improve port and rack utilization, and relax power and space constraints on important routes. Those benefits are economically meaningful, but they are only 1 component of hyperscaler AI unit economics, where the dominant cost pools remain compute, power delivery, cooling, and real estate. The relevant impact is best understood as better monetization of constrained network assets and lower cost to move bits across critical routes, not as an immediate or easily quantifiable lift to hyperscaler consolidated margins.  ROUTE DIVERSITY AND GEOPOLITICAL IMPLICATIONS The post is strongest on the commercial value of route diversity. Current Ciena and Lightstorm materials explicitly market JGA as a low-latency alternative to South China Sea transit with operational and geopolitical diversity, and Lightstorm positions the route for cloud, AI, gaming, and financial-services use cases where low latency and resilience command a premium. That commercial framing is credible. Carnegie documents that Chinese claims and permitting requirements delayed SJC2 by several years and contributed to rerouting of Apricot and Echo around Indonesia. CSIS documents growing approval friction and rerouting away from vulnerable areas, while reporting from The Washington Post describes repair delays measured in months, projects delayed by years, and materially higher development costs for routes that avoid the South China Sea. The original Google framing of JGA in 2018 emphasized scalable, diverse, low-latency APAC connectivity rather than geopolitics alone, so the phrase “purposefully avoids the South China Sea” is somewhat more absolute than the historical record supports. The economically relevant point is that JGA now occupies a premium route position where latency, resilience, and geopolitical optionality are simultaneously monetizable.  IMPLICATIONS FOR CIENA OPTICAL NETWORKING Immediate revenue materiality should not be overstated. No contract value, booking size, or margin profile was disclosed for the Lightstorm deployment, so the event cannot be assumed to move near-term earnings materially against Ciena’s $1.43B quarterly revenue base. The importance lies in the pattern. Ciena reported Q1 FY2026 direct cloud-provider revenue up 76% YoY to 42% of total revenue, non-telco at 53% of total revenue, and optical revenue up more than 40% YoY, led by Waveserver and RLS, each up more than 80% YoY. Management also disclosed that 3 hyperscalers are using Ciena optical solutions for training applications across distance and described RLS as the de facto cloud-provider line-system standard. The Lightstorm event therefore matters less as a stand-alone revenue item and more as a verified reference point inside a broader cloud and AI optical upcycle.  The post’s statement that this validates Ciena’s pivot to a critical AI infrastructure vendor is directionally understandable but analytically imprecise. Ciena has not pivoted away from optical transport. It remains fundamentally a high-speed optical and connectivity vendor whose customer mix and demand vectors are shifting sharply toward cloud and AI. Management has explicitly described Ciena as a critical provider of the high-speed optical systems and interconnects that enable AI workloads to scale and be monetized, and has said the company is taking meaningful share of AI-driven connectivity spend. Dell’Oro ranks Ciena among the top 3 DCI suppliers by revenue share, alongside Nokia and Cisco. The correct investment framework is role expansion and TAM broadening, not business-model reinvention.  Visibility has improved, but the post overstates certainty. Ciena raised FY2026 revenue guidance to $5.9B to $6.3B, or 28% growth at the midpoint, and management cited strong order book, record backlog, and multi-year AI-driven demand. However, the 10-K still states that results can fluctuate materially, that some revenue is generated on a book-to-revenue basis within the same quarter, that purchase orders may be modified or canceled, and that AI-driven adjacent markets are nascent or dynamic and difficult to forecast. Customer concentration remains material. Ciena disclosed that its top 5 customers represented 49.7% of FY2025 revenue and that 1 cloud provider represented 17.9% of FY2025 revenue. The appropriate conclusion is improved medium-term visibility with substantial timing and concentration risk, not annuity-like predictability.  IMPLICATIONS FOR THE GENERATIVE AI ECOSYSTEM For the generative AI ecosystem, the most important implication is that networking bottlenecks are propagating outward from the cluster into the campus, metro, regional, and subsea layers. Ciena has stated that as AI data centers hit power and space limits, customers are distributing compute across multiple sites and using high-speed optical networks to create a single training environment across distance. LightCounting states that AI clusters no longer fit into a single building because electricity constraints force multi-location GPU deployments. Meta has publicly described a 1 GW AI cluster spanning multiple data center buildings, weatherproof tents, and adjacent colocation facilities, while evolving software to support long-distance training across geographically distributed sites. In that context, JGA is not merely a telecom transport upgrade. It is evidence that WAN and subsea transport are increasingly part of the AI system boundary.  That said, the Lightstorm and JGA development sits at the wide-area and front-end edge of the AI network stack, not at the most performance-sensitive frontier of back-end training fabrics. Dell’Oro reports that Ethernet now accounts for more than 2/3 of AI back-end switch sales and that most 800G switch shipments remain concentrated in back-end networks, while 100/200/400G still account for more than 2/3 of front-end switch sales. Subsea 400G therefore matters for data mobility, redundancy, cross-region inference, model distribution, checkpoint movement, and global cloud backbone performance, but it does not replace the economics or performance bottlenecks inside GPU clusters. The correct read-through is that AI demand is broadening the optical opportunity across the network hierarchy, not collapsing all AI networking layers into 1 homogeneous market.  The medium-term significance lies in the upgrade path. Ciena stated that JGA can scale from 400G to 800G in the near future, and Ciena and Meta subsequently demonstrated 800G over 16,608 km on the transpacific Bifrost cable, reaching 18 Tb/s per fiber pair with a 50% reduction in watts per bit versus the prior generation. That matters because the long-run value in subsea is not 1 discrete 400G turn-up. It is the repeatability of terminal, line-system, coherent DSP, and control-plane refresh cycles on scarce, strategic routes as cloud replication, AI training distribution, and global inference traffic continue to rise. The Lightstorm event is therefore better understood as an early commercial waypoint on a broader subsea capacity migration path rather than as an endpoint.  INVESTMENT CONCLUSION Overall, the post is materially correct on the core fact pattern and on the broad direction of the Ciena and cloud-optical thesis. It is overstated on 3 dimensions: the precision of the late-2025 400G economic crossover claim, the universality of the roughly 40% power-reduction figure, and the certainty implied by “direct margin expansion,” “highly visible revenue runways,” and “secure growth run.” The most defensible investment conclusion is that the JGA upgrade is a verified proof point that AI and cloud demand are now influencing subsea and DCI refresh cycles, that Ciena is a meaningful beneficiary of that spend, and that geopolitical route diversity is becoming a monetizable infrastructure attribute. The least defensible conclusion is that 1 deployment materially de-risks Ciena’s earnings model or independently justifies a valuation rerating without continued evidence of repeat wins, sustained cloud mix, and durable margin capture across the broader cloud and AI optical portfolio. 
Weekend is a great time to do some research. Companies I’m going to look into this weekend: $AEHR $FN $CIEN $JBL $VLN $MTSI $IRDM $SPIR Here is my method 👇
This is another clear bullish signal for scale-out optical networking and $CIEN. As AI workloads continue to scale, the bottleneck is no longer just compute—it is increasingly the ability to move data efficiently across distributed infrastructure. Hyperscalers and model providers are being forced to rethink architecture, shifting from tightly coupled, single-site clusters toward geographically distributed compute fabrics. This shift is being driven by both power constraints and demand variability, requiring workloads to be dynamically balanced across regions rather than concentrated in a single location. That architectural evolution materially increases the importance of high-capacity, low-latency optical interconnects over longer distances. In a world where training and inference jobs can be scheduled across metros—or even inter-regionally—the network effectively becomes the backplane of the AI system. This is where scale-out optical networking becomes critical. For Ciena, this trend directly expands the addressable market. The company is positioned at the intersection of coherent optics, routing, and automation—exactly the layers required to enable this distributed compute paradigm. As bandwidth demand accelerates and network architectures flatten, spend shifts from traditional datacenter switching toward transport and optical layers that can handle massive east-west traffic across distance. Importantly, this is not a cyclical upgrade—it is a structural change in how compute infrastructure is built. The need to orchestrate AI workloads across distributed clusters introduces persistent demand for higher-capacity wavelengths, more flexible optical systems, and tighter integration between compute and network layers. In effect, AI is turning the WAN into an extension of the datacenter. That is a durable tailwind for optical networking vendors, and particularly for players like Ciena that are deeply embedded in hyperscaler and carrier ecosystems.
$LITE $COHR $CIEN $AAOI EXECUTIVE OVERVIEW Across the LightCounting materials, the common thesis is that optical connectivity has moved from a supporting component to a primary scaling constraint inside AI infrastructure. LightCounting estimates total sales of optical transceivers and related products reached $23.8 billion in 2025, with Ethernet transceivers near $18 billion and AOCs above $1.1 billion. Separately, its January 2026 AI optics note puts the combined market for Ethernet optical transceivers and CPO used across AI and non-AI applications at $16.5 billion in 2025 and $26 billion in 2026, both implying 60% growth, while optics going to the top 5 cloud companies rises from 2.7% of capex in 2025 to 3.1% in 2026 and 4.1% by 2031. The important implication is not only that cloud capex is rising, but that the optical content intensity of each incremental AI dollar is rising as architectures become more communication-heavy. This is not a simple straight-line supercycle argument. LightCounting repeatedly warns that capacity additions in lasers and modules are catching up, that flat quarters could appear in late 2026 as supply and demand rebalance, and that the optical market still carries the historical pattern of 2-3 strong years followed by flat or negative periods. The March 2026 Ethernet Optics note explicitly frames 2027-2031 through 3 scenarios: a “Soft Landing,” a “Bumpy Ride,” and a “Fantastic Growth” case that could push annual AI-cluster optics toward $100 billion by 2030, but only if macro stability, meaningful AI productivity gains, strong scale-up optical adoption, and limited high-speed copper substitution all align. The table on page 5 of that note makes the framework explicit by listing economic cycles, model and cluster efficiency gains, TPU substitution, training-to-inference mix shift, and copper progress as swing variables. The February capex note is even more cautionary, arguing that the top 5 cloud companies represented 68% of total capex in 2025, plan to nearly double spending in 2026, and could still trigger an avalanche through the supply chain if even 1 customer moderates spend. The correct inference is structurally bullish on optics but tactically skeptical of straight-line earnings extrapolation into 2027. The most important architectural shift in the LightCounting set is that optics is expanding along 3 separate axes at once: scale-out, scale-up, and scale-across. Scale-out remains the volume engine and still leans heavily on standardized pluggable Ethernet modules. Scale-up is the major new content unlock, because LightCounting argues it needs almost 10x the bandwidth of scale-out and may become the decisive use case for CPO and NPO as clusters move from a single rack toward 4-8 rack domains. Scale-across is the later but potentially longer-duration transport opportunity, with LightCounting estimating roughly 20% of data-center DWDM traffic in 2030 will be tied to AI scale-across architectures and citing scenarios that require 5-50 Pbps and hundreds to thousands of fiber pairs in C+L bands. The chart on page 3 of the January 2026 AI note visually shows AI scale-out dominating current spend while AI scale-up becomes the incremental growth wedge later in the forecast. That 3-axis framework materially broadens the beneficiary list beyond conventional transceiver vendors. WHAT THE LIGHTCOUNTING REPORTS ARE REALLY SAYING CPO is clearly past the concept stage, but the reports do not support the simplistic view that pluggables are about to be displaced wholesale. LightCounting’s December 2025 CPO note says Nvidia introduced 200G/lane CPO on InfiniBand and Ethernet switches in March 2025, Meta provided strong reliability data on Broadcom CPO, Broadcom launched 3rd-generation 200G/lane CPO, and Nvidia reported 10x higher resiliency and 5x higher cluster uptime versus pluggables. At the same time, the January 2026 “Waiting for Green Light” note says customer acceptance remains the main gating issue and that early deployments are likely to begin in smaller, risk-tolerant clusters before broad top-5 cloud adoption. Nvidia’s own roadmap reinforces the staged nature of the transition: Spectrum-X and Quantum-X photonics are in market for rack-to-rack switching, but the current Rubin NVL72 still uses a copper NVLink spine inside the rack. Even the stacked chart on page 5 of the December 2025 CPO note still leaves transceivers as the largest 2030 bucket, with cables still substantial and CPO additive rather than instantly dominant. The implication is that 2026-2027 remains a hybrid era, with pluggables, AECs, ACCs, AOCs, LRO/LPO, NPO and CPO all coexisting by reach and failure-cost profile rather than 1 form factor immediately winning the entire stack. (NVIDIA Newsroom) The standardization wave at OFC 2026 is therefore strategically important. LightCounting highlighted the OCI MSA and the CPX MSA as attempts to pull the industry out of a proprietary “wild west,” while Arista simultaneously launched the XPO MSA as a dense liquid-cooled pluggable alternative. Broadcom described OCI as an open specification for low-power optical SerDes in multi-rack scale-up systems, Ciena’s CPX and Vesta 200 offer a pluggable co-packaged optical engine based on the Nubis architecture, and Arista’s XPO targets 12.8 Tbps per module with 4x the front-panel density of 1600G OSFP while preserving serviceability. The strategic implication is that ecosystem opening should accelerate adoption but also redistribute value away from proprietary system lock-in toward differentiated component suppliers, thermal and control software, test vendors, and companies whose architectures remain field serviceable. That is positive for Lumentum, Coherent, Ciena, Viavi, and Aehr in different ways, and less favorable for any thesis that assumes 1 closed CPO stack captures the entire value pool. China is an important incremental swing factor, but not a clean visibility source. LightCounting argues that demand from Chinese cloud companies has reaccelerated sharply, with the U.S.-China optics spending gap narrowing from 7x in 2022 to 4x in 2024-2025 and potentially 2.5x by 2031, with 2025-2031 CAGR of 18% in the U.S. versus 29% in China. That is materially supportive for suppliers with China exposure and for the global supply chain at 800G, 1.6T and eventually 3.2T. However, the external policy environment remained volatile even after LightCounting’s note: Reuters reported conditional U.S. approval for some Nvidia H200 exports to China in January 2026, but also reported Chinese customs restrictions, approval frictions, and renewed U.S. debate over AI-chip export rules. The correct conclusion is that China adds meaningful volume upside, but not dependable quarter-to-quarter predictability. (Reuters) The supply-side structure of the optics market matters as much as demand. LightCounting estimates optical transceiver and related-product sales rose 55% in 2025, notes that Innolight reached $1.87 billion of quarterly revenue in Q4 2025, estimates Eoptolink above $1 billion, and says Coherent, Fabrinet, and Lumentum posted record levels as well. Yet the same March 2026 note warns that optical chip and module capacity is now catching up with demand, which could intensify supplier competition and drive sharper price declines by the end of 2026. This means stock selection should increasingly favor constrained technology layers rather than commoditizable assembly. In practical terms, upstream lasers, InP, silicon photonics content, optical switching, and manufacturing and test infrastructure should prove more defensible than merchant module assembly alone if the market transitions from shortage economics to competition economics. IMPLICATIONS FOR OPTICAL NETWORKING AND THE GENERATIVE AI INFRASTRUCTURE BUILD-OUT Optical networking is no longer a back-end utility. It is becoming a first-order determinant of AI-system economics because networking increasingly shapes cluster size, GPU utilization, uptime, power efficiency, and rack-to-rack or campus-to-campus topology. That is the meaning of the OFC 2026 formulation that optics is now “inside the machine.” Once that framing is accepted, value creation shifts away from the old question of how many pluggable ports are shipped and toward a broader question of which companies own the laser sources, photonic engines, switching layers, software control, and validation tools that make larger AI systems economically operable. This is why the LightCounting notes are not merely bullish for transceivers; they are bullish for the entire photonic control stack. Optical circuit switching is one of the most underappreciated beneficiaries of this transition. Once optics moves closer to the compute complex, the problem is no longer only bandwidth density; it is also utilization, failure isolation, and dynamic topology management. Lumentum disclosed in February 2026 that OCS backlog was already above $400 million and paired that with an incremental multi-hundred-million-dollar CPO order for 1H 2027. Coherent’s OFC materials describe OCS as a new growth engine shipping into production deployments with more than 10 customer engagements, and Marvell demonstrated interoperability between its optical DSPs and Lumentum’s R300 OCS for AI scale-up fabrics. OCS therefore appears to be emerging as an architecture layer, not a niche adjunct, especially if AI fabrics require reconfigurable optical paths to improve accelerator utilization and resiliency. (Lumentum Investor Relations) LightCounting’s DSP note also reframes how value should be allocated across semiconductors. In 2025, 800G PAM4 chipset shipments nearly tripled and 1.6T chipset sales are expected to exceed $2 billion in 2026, but LightCounting also expects PAM4 growth to moderate in 2027-2031 as linear-drive optics and CPO displace some retimed DSP content. Conversely, coherent DSP growth should reaccelerate as coherent-lite and inter-building AI training traffic expand. That is a critical nuance: pure retimer and DSP beta is likely strongest in the current pluggable ramp, while the longer-duration upside migrates toward a broader set of optical components and coherent-lite transport solutions. The companies best positioned are those that can monetize both phases rather than only 1. COMPANY IMPLICATIONS LUMENTUM Lumentum is one of the best-aligned public equities for the LightCounting thesis because it sits in multiple profit pools that expand as the optical stack thickens. Near term, it remains exposed to the 800G and 1.6T pluggable cycle through components and modules. Medium term, it is positioned in the highest-value CPO inputs through CW and ultra-high-power lasers, external laser source modules, and optical circuit switches. Longer term, it also has scale-across exposure through coherent and transport optics, and it has introduced a 1060 nm VCSEL-based scale-up platform that deliberately provides an alternative supply chain to silicon photonics and InP-laser CPO approaches. Operationally, Q2 FY26 revenue was $665.5 million, components were 66.7% of revenue, OCS backlog was already above $400 million, management disclosed an incremental multi-hundred-million-dollar CPO order for 1H 2027, and the company announced a new 240,000 square foot U.S. InP laser facility in North Carolina with Nvidia as a customer and a mid-2028 production ramp. The read-through is that Lumentum is not merely a transceiver beneficiary; it is a control-point supplier to the system bottlenecks that LightCounting thinks become more important as scale-up and CPO adoption deepen. The main risk is that this makes the company highly sensitive to customer concentration and capacity timing if 2027 becomes a rebalancing year rather than a continuation year. (Lumentum Investor Relations) COHERENT Coherent appears to have the broadest portfolio alignment with the full LightCounting framework. The company’s portfolio maps directly onto the 3 scaling vectors: pluggable transceivers and laser content for scale-out, CPO and NPO building blocks and high-power CW and ELS for scale-up, and DCI transceivers, transport optics, multi-rail, and optical circuit switching for scale-across. Coherent’s official OFC 2026 materials describe CPO and NPO revenue starting in H2 2026, position OCS and multi-rail as incremental SAM expansion, and show a deep datacom stack spanning VCSELs, EMLs, CW lasers, silicon photonics, detectors, thermoelectric coolers, external laser sources, and passive optics. Coherent also disclosed that InP capacity is doubling across 2026 and 2027, that it is ramping 6-inch InP in Sherman, Texas for high-power CW lasers and ELS, and that its technology leadership extends to 100G, 200G, and 400G EMLs for 800G through 6.4T. Financially, Coherent reported Q2 FY26 revenue of $1.69 billion and highlighted strong data-center and communications demand with continued growth expected into FY27. Relative to most peers, Coherent offers the cleanest hedge against architecture uncertainty: its exposure remains favorable whether pluggables remain dominant longer, CPO and NPO ramp sooner, or AI networking extends into DCI and transport. The risk is not architectural obsolescence; it is margin compression if the industry overbuilds photonic capacity while hyperscaler ordering volatility rises. (Coherent Inc) CIENA Ciena is one of the biggest secondary winners from the LightCounting work because the reports materially enlarge the addressable market around, between, and increasingly inside AI data centers. On the inside-DC side, Ciena’s acquisition of Nubis in September 2025 and launch of the Vesta 200 6.4T CPX pluggable optical engine in February 2026 give it a credible position in the packaged-optics debate without forcing it into a single proprietary CPO architecture. On the around-and-between-DC side, Ciena is already highly leveraged to AI-driven DCI and scale-across demand, which LightCounting expects to be a meaningful share of DWDM traffic by 2030. This is already visible in operating data: Ciena said direct cloud provider revenue grew 76% y/y and represented 42% of revenue in Q1 FY26, while revenue and shipments of RLS grew more than 80% y/y, fueled by AI-driven cloud expansion. The investment implication is that Ciena may not show the same immediate torque as pure-play 800G and 1.6T merchants in 2026, but it likely has better duration if the optical spend mix continues shifting from in-building connectivity toward coherent-lite, DCI, and scale-across architectures. That makes it more of a compounding network-systems beneficiary than a short-cycle optics squeeze name. (Ciena Corporation) VIAVI Viavi is a beneficiary of complexity rather than of port count. That distinction matters because the LightCounting materials increasingly frame AI networking as a reliability, interoperability, and serviceability problem, not only a raw-bandwidth problem. As the stack moves from 800G to 1.6T and toward multivendor AI fabrics with LPO, NPO, CPO, and optical scale-up, qualification intensity and failure-cost asymmetry both rise. Viavi’s March 2026 launch of its TestCenter D2 1.6T appliance explicitly targets hyperscalers, neoclouds, and network equipment manufacturers, and the company stated that the shift from 800G to 1.6T deployments for AI back-end infrastructure is expected to begin in earnest in 2026. Its October 2025 acquisition of Spirent’s high-speed Ethernet, network-security, and channel-emulation testing business also materially broadened its AI data-center testing stack. The implication is that Viavi is a relatively lower-beta but higher-quality way to express the thesis that AI optics is becoming harder to validate, harder to debug, and more economically costly to fail. (VIAVI Solutions Inc.) AEHR TEST SYSTEMS Aehr is the most asymmetric beneficiary if the LightCounting vision of packaged optics and optical I/O expands deeper into the compute fabric. The core reason is that Aehr monetizes the reliability and early-life-failure screening problem at wafer level, before photonic devices are packaged into expensive modules or co-packaged systems. In March 2026 Aehr disclosed a follow-on order from a lead silicon photonics customer for systems used in data-center optical interconnects and emerging optical I/O architectures, with each system able to test up to 9 300 mm wafers in parallel at up to 3500 W per wafer. It also announced a new major silicon photonics customer buying systems for both engineering qualification and high-volume production, and explicitly said silicon photonics could become a meaningful long-term growth driver for its wafer-level burn-in business. The LightCounting implication is straightforward: as optics moves closer to the ASIC and the serviceability penalty of field failures rises, burn-in and reliability screening become more valuable, not less. The main uncertainty is technology mix. If the industry stays longer in discrete-laser pluggables or shifts meaningfully toward VCSEL-based scale-up alternatives, Aehr’s ramp can be slower than the broad AI-optics narrative implies. (Aehr Test Systems) AXT AXT is a genuine upstream read-through to the LightCounting thesis, but not a clean one. LightCounting’s repeated emphasis on InP laser demand, easing shortages, and capacity additions directly supports InP substrate demand, and AXT itself said in February 2026 that Q1 growth should be driven primarily by indium phosphide for the AI infrastructure build-out, that it is broadening exposure to Tier-1 customers, and that it is on track to double indium phosphide manufacturing capacity in 2026. The company also identifies data-center connectivity and silicon photonics as end markets for its substrates. The positive case is that an AI-optics cycle increasingly built on EMLs, CW lasers, and silicon photonics light sources should raise InP substrate demand over several years. The negative case is that AXT’s manufacturing and permit exposure in China adds a separate volatility layer that can dominate near-term results even when end demand is healthy. AXT therefore fits best as a leveraged but geopolitically noisy input supplier to the optics build-out. (AXT Investor Relations) AAOI AAOI is one of the clearest near-term beneficiaries of what LightCounting is actually forecasting, because the forecast still expects 2026 to be dominated by merchant pluggables, active cables, and 1.6T ramps rather than by immediate wholesale CPO substitution. AAOI reported Q4 2025 data-center revenue of $74.9 million, up 69% y/y and 70% sequentially, received its first 800G volume order from a major hyperscale customer, and guided to more than $1 billion of revenue in 2026. In March 2026 the company announced its first 1.6T data-center transceiver volume order from a major hyperscale customer worth more than $200 million, with shipments starting in Q3 2026, and said it expects to reach combined 800G and 1.6T production above 500,000 units per month by year-end. It has also expanded U.S. and Taiwan production, highlighted 10M+ pieces of annual 100G, 400G, 800G and 1.6T capacity, and is sampling NPO, OBO and ELSFP products while building a large Texas AI-optics facility. The implication is that AAOI can work very well if the next 12-18 months are about hyperscaler pluggable ramps, U.S. sourcing, and speed transitions from 800G to 1.6T. The risk is that its moat is narrower than upstream laser or system-control names, so the late-2026 price-down phase that LightCounting warns about could hit AAOI harder than more differentiated peers. (Applied Optoelectronics, Inc.) OTHER IMPORTANT READ-THROUGHS Credo and Marvell are important additional beneficiaries because LightCounting’s base case is still extremely favorable for the bridge technologies between conventional pluggables and full CPO. LightCounting explicitly raised forecasts for ACCs and AECs, still expects fully retimed transceivers to drive the largest chipset dollar growth in the current phase, and only later expects LPO and CPO volume deployments to pressure PAM4 DSP sales. That framework lines up well with Credo’s 1.6T LRO, ZeroFlap optics and AEC portfolio, and with Marvell’s mass-volume Ara 1.6T DSP platform and broader end-to-end AI-connectivity stack. Marvell is especially well hedged because it spans both the current DSP-heavy pluggable phase and the later scale-up photonics phase, including optical DSPs, coherent-lite, and interoperability with Lumentum OCS. (Credo Technology Group) Broadcom and Nvidia set the roadmap even when they are not the purest investment vehicles for optics. Broadcom’s 3rd-generation 200G/lane CPO is explicitly targeted at scale-up domains beyond 512 nodes, while Nvidia’s silicon photonics switching claims 5x better network power efficiency, 10x higher resiliency, and 5x longer sustained AI application runtime versus pluggable-transceiver-based networks. Those roadmaps support LightCounting’s argument that optics is moving into the heart of the AI system. At the same time, Nvidia’s current rack-scale Rubin platform still uses a copper NVLink spine inside the rack, which underscores that copper remains viable at the shortest reaches and that the opticalization of scale-up is a progression, not a singular event. That nuance is critical when comparing higher-beta pluggable names to longer-duration CPO, NPO and optical-switching beneficiaries. (Broadcom Inc.) Arista and Cisco should also be interpreted as evidence that the pluggable ecosystem is adapting aggressively rather than ceding the field to monolithic CPO. LightCounting’s January 2026 note framed Arista and Google as not planning CPO at that point, but by March 2026 Arista had launched the XPO MSA, a dense liquid-cooled pluggable architecture for AI fabrics, while Cisco was emphasizing 1.6T OSFP optics and energy-efficient AI-networking systems built around Silicon One G300. That supports the view that the end-state is likely segmented by reach, serviceability, cooling, and operator preference rather than determined by a single universal form factor. Companies tied exclusively to 1 architecture should therefore trade at a discount to companies with credible exposure across several form factors. (Arista Networks) The module market is likely to remain intensely competitive because Chinese vendors continue to scale quickly. LightCounting highlighted record or near-record performance at Innolight, Eoptolink, Accelink, Hisense, and HG Genuine, while also forecasting a sharp reacceleration in Chinese cloud optics demand. This matters for all U.S. names. The most durable winners are unlikely to be those with the highest gross number of shipped pluggables; they are more likely to be the companies that own scarce upstream content, differentiated optical engines, optical switching, or validation infrastructure. That is the central reason the LightCounting work is more structurally bullish for Lumentum, Coherent, Ciena, Viavi, and Aehr than for simple merchant-assembly stories, even if the latter can outperform tactically during shortage phases. BOTTOM LINE The highest-conviction inference from the LightCounting set is that optical networking is no longer a derivative of compute; it is becoming 1 of the primary determinants of AI-system architecture, power efficiency, uptime, and therefore monetization. The best-positioned public companies are the ones with exposure to multiple scaling layers and multiple form factors. In that framework, Coherent and Lumentum appear best aligned to the broadest range of outcomes, Ciena appears best positioned for the longer-duration scale-across and open-packaged-optics opportunity, Viavi and Aehr benefit from rising qualification complexity, AAOI offers the highest near-term beta to a still-pluggable 2026 cycle, and AXT is a higher-risk upstream material beneficiary. The most important risk factor is not whether AI spend remains large in 2026; it is whether the market mistakes a structural optical-intensity trend for a straight-line annual earnings trend and ignores the very real rebalancing and pricing risks that LightCounting flags for late 2026 and 2027. (Lumentum Investor Relations) Longer term, the SC25 note is also worth attention because it places AI within a broader convergence of HPC, quantum, and optical computing, and it reported SCinet peak traffic of 13.72 Tbps versus 8.41 Tbps the prior year. That does not change 2026 earnings models, but it reinforces the strategic direction: the industry is moving toward a world in which photonics is not only the fabric between machines, but increasingly part of the compute-adjacent substrate itself. If that direction proves correct, the long-duration winners will be the companies that own photonic building blocks, interoperability layers, and manufacturing know-how rather than only the current generation of pluggable module SKUs.
Strong follow-on day for $CIEN . Great to see it being picked up by international b/ds now, and broadening the shareholder base. Still feels cheap here. I am long, and I don't see that changing any time soon. https://t.co/MhmYctmZa6
$LITE $COHR $CIEN $AAOI AS OF MARCH 30, 2026 EXECUTIVE OVERVIEW The March 30 drawdown in AI photonics and optical infrastructure is best understood as a policy-scare de-risking of a crowded AI-optics complex rather than a clean repricing of near-term earnings power across the full value chain. The policy catalyst is real: the January 14 Section 232 semiconductor proclamation established a 2-phase framework, imposed an immediate 25% tariff on a very narrow class of advanced computing chips, and required Commerce and USTR to provide a 90-day update by April 14. But the market move implies a much broader and much more immediate tariff regime than is currently in force. The operative question is not whether risk exists, but whether current price action matches current legal scope. On that test, the selloff appears directionally rational at the most exposed tail of the basket and excessively broad across the middle.  ASSESSMENT OF THE SOURCE CLAIMS The source framing is directionally correct on 3 points. AAOI and AXTI are among the most exposed names to trade and supply-chain friction. The April 14 checkpoint is a legitimate event risk for the group. Domestically anchored photonics platforms can plausibly outperform in a tariff scare. However, 4 elements of the framing require tightening. April 14 is an update deadline on the status or outcome of negotiations, not an automatic broad-tariff implementation date. The current 25% duty is targeted at a narrow set of advanced AI chips defined through specific technical thresholds, not the full optical or photonics stack. Taiwan exposure should not be treated as equivalent to China exposure because Washington and Taipei already reached a semiconductor-focused trade arrangement in January and finalized a broader tariff deal in February, with Taiwan stating in late February that preferential semiconductor treatment would not change. The characterization of POET as pre-revenue is also too strong; revenue remains very small, but NRE and product revenue were reported in 2025 and an initial production-volume optical engine order was disclosed in September 2025.  POLICY MECHANICS The January 14 proclamation is more nuanced than the tape suggests. Clause 7 establishes a 2-phase approach. Phase 1 consists of continued negotiations with foreign jurisdictions that can strengthen the U.S. semiconductor industry plus an immediate 25% tariff on a very narrow category of chips important to the administration’s AI policy. Phase 2 contemplates broader semiconductor tariffs after negotiations conclude, potentially paired with a tariff-offset program for companies investing in U.S. semiconductor production and supply-chain capacity. The proclamation then requires Commerce and USTR to update the President within 90 days and explicitly preserves authority for later action if agreements are not concluded or are ineffective. That means April 14 is an information event and policy checkpoint; it is not, on the face of the proclamation, the only decision date that matters, and it is not a self-executing broad tariff deadline. The 180-day framework embedded in the proclamation implies that July 13 is also relevant for the evolution of the regime.  The CBP implementation guidance further narrows the immediate economic impact. The new 25% duty applies to certain semiconductor articles and derivative products classified in a limited set of HTSUS provisions and meeting very specific total processing performance and DRAM bandwidth thresholds consistent with H200- and MI325X-class advanced AI chips. CBP simultaneously created 0% treatment for non-threshold products and for multiple end uses, including U.S. data centers, repairs, R&D, startups, non-data-center consumer electronics, non-data-center civil industrial applications, and public-sector uses. The proclamation also prevents covered products from stacking with certain other tariff programs. In practical terms, the current phase-1 regime looks engineered to tax a narrow class of imported leading-edge AI accelerators while minimizing disruption to domestic AI infrastructure buildout. That sharply limits the immediate direct relevance to optical modules, photonic materials, and much of the equipment and test stack on March 30.  MARKET CONTEXT AND POSITIONING The broader market backdrop also matters. On March 30, U.S. equities were already under pressure from a widening Middle East conflict, higher oil prices, and renewed inflation concerns. The S&P 500 fell 0.39%, the Nasdaq fell 0.73%, and the semiconductor index fell 4.2%. Against that background, the photonics basket traded with both sector beta and crowding pressure. Lumentum and Coherent had entered March 30 after Reuters-reported positive catalysts from Nvidia’s March 2 decision to invest $2 billion in each company to bolster U.S. photonics capacity and from their March 6 addition to the S&P 500 effective March 23. The most plausible interpretation is that the market used a real tariff headline to aggressively de-risk 1 of the highest-multiple, highest-momentum segments inside AI infrastructure. The result looks less like a clean fundamental rerating across every layer of the value chain and more like a forced compression of crowding, duration, and policy uncertainty into a single session.  CROSS-SECTIONAL MESSAGE The tape was broad, but it was not truly indiscriminate. AAOI closed down 13.24%, AXTI 12.95%, COHR 9.84%, CIEN 9.09%, TSEM 7.83%, MRVL 7.45%, LITE 6.93%, GFS 3.61%, GLW 6.06%, FN 10.98%, while ALMU rose 4.10%. That spread is consistent with a market distinguishing, albeit imperfectly, between direct tariff and export-control exposure, high-beta sympathy risk, and domestic-manufacturing optionality. The highest-conviction losers were concentrated in names with meaningful China exposure, substantial Taiwan dependency, or elevated AI-optics momentum. The relative winner was the most clearly domestic and early-stage photonics platform in the set. OPTICAL TRANSCEIVERS AND ACTIVE NETWORKING AAOI is the clearest case where the source note and the filings align. The company explicitly reports manufacturing and R&D facilities in the U.S., Taiwan, and China. It discloses that 57.5% of 2025 revenue was manufactured at its China-based subsidiary and 38.2% at its Taiwan-based facility. Its filings also state that a significant portion of manufacturing operations are based in Ningbo, China and that new tariffs, legislation, regulations, or executive orders could have material adverse effects. Laser chips and optical components are generally manufactured in Sugar Land, while optical components, subassemblies, and optical equipment are manufactured in Taiwan and China. That makes AAOI unusually exposed to any broadening of U.S. tariff scope, any retaliatory action, and any incremental friction in China-based assembly. The 13.24% decline is therefore among the more fundamentally grounded moves in the basket. The caveat is that even AAOI is not yet being directly targeted by the current narrow Section 232 phase-1 duty; the stock is discounting a phase-2 scenario, not reacting to a current transceiver tariff.  The same framework does not apply equally to every optical name. Ciena relies on contract manufacturers with facilities in Canada, Mexico, Thailand, Vietnam, and the U.S., and acknowledges reliance on a small number of contract manufacturers for the majority of production. That introduces manufacturing concentration and supply-chain risk, but it is not the same China and Taiwan import-risk profile as AAOI. Fabrinet is another example of over-bundling by the market. Its 2025 10-K states that the majority of its assets and manufacturing operations are located in Thailand, with the remaining facility footprint in China, the U.S., Israel, and the Cayman Islands. Fabrinet is clearly exposed to trade-policy volatility, but the dominant operational concentration is Thailand rather than Taiwan, and the March 30 move therefore looks more like optical-basket contagion than a precise read-through from the Section 232 checkpoint. Ciena’s 9.09% drop and Fabrinet’s 10.98% drop appear materially less justified by current direct tariff mechanics than AAOI’s decline.  Lumentum and Coherent sit in an intermediate bucket. Lumentum continues to invest heavily in next-generation optical components and transceiver modules for cloud, AI/ML, and data-center interconnect applications. It manufactures through a mix of internal fabs and assembly and test sites, with significant facilities in the U.S., Thailand, China, the U.K., Slovenia, and Japan. It also disclosed that trade policy and tariffs have affected and may continue to affect net revenue and that some product lines were moved out of China due to U.S. export restrictions. Coherent is even more geographically diversified, with principal non-U.S. production and R&D in China, Finland, Germany, Malaysia, the Philippines, Singapore, South Korea, Sweden, Switzerland, the U.K., and Vietnam. Coherent stated that tariffs and trade sanctions between the U.S. and China did not have a material impact in fiscal 2025 and are not expected to have a material impact in fiscal 2026, citing its diversified manufacturing footprint. Against that backdrop, Lumentum’s 6.93% decline and Coherent’s 9.84% decline look too severe if the analytical frame is current legal tariff incidence, but not unreasonable if the market is de-rating crowded AI-optics winners on the probability of a broader phase-2 regime.  NETWORKING SILICON Marvell is 1 of the more important names in the basket because it bridges optical demand and advanced-node semiconductor dependency. The company states that most of its products are manufactured by 3rd-party foundries in Taiwan, with other sources in China, Germany, South Korea, Singapore, and the U.S. Most assembly, testing, and packaging capacity is in China, Malaysia, Singapore, Taiwan, and Canada. Marvell also states that the April 14, 2025 BIS Section 232 investigation covered semiconductors, semiconductor manufacturing equipment, substrates, bare wafers, legacy chips, leading-edge chips, microelectronics, and other components, and that the result could be additional tariffs and trade restrictions. In addition, Marvell carries $2.67 billion of unconditional purchase commitments with foundries and test-and-assembly partners. That combination of advanced-node concentration, limited alternative foundries, Asia-heavy back-end operations, and fixed supply commitments makes Marvell 1 of the more sensitive names if phase 2 broadens. The 7.45% decline is therefore intellectually defensible as scenario pricing. But the current phase-1 tariff regime remains far narrower than the market is discounting.  COMPOUND SEMICONDUCTOR SUBSTRATES AND FOUNDRIES AXTI may be the purest expression of real policy exposure in the group. The company’s investor materials state that it has Asia headquarters in Beijing and manufacturing facilities in 3 separate locations in China. In January 2026, AXT cut its 4Q25 revenue outlook primarily because fewer export-control permits for indium phosphide were being issued by China’s Ministry of Commerce than previously expected. Because indium phosphide is a critical material for transceiver lasers and broader photonics applications, the stock is exposed not only to potential U.S. tariff escalation but also to already-existing Chinese export-control friction. That makes AXTI’s 12.95% decline 1 of the more justified moves in the entire group. The market is correctly treating AXTI as a direct picks-and-shovels casualty of escalating localization and trade-friction risk.  GlobalFoundries and Tower should not be placed in the same bucket as AAOI or AXTI. GlobalFoundries operates 4 manufacturing sites in Dresden, Singapore, Malta, New York, and Burlington, Vermont, and explicitly markets itself as the only scaled pure-play foundry with a global footprint not based in China or Taiwan. Tower operates 1 facility in Israel, 2 in the U.S., 2 in Japan, shares a 300mm facility in Italy, and has access to an Intel capacity corridor in New Mexico. Both companies have real semiconductor-cycle exposure and both could be affected by broader tariffs on derivatives or by customer caution. But the direct supply-chain vulnerability embedded in current policy is materially lower than for China-heavy or Taiwan-dependent peers. In a true phase-2 localization regime, both companies could become relative beneficiaries rather than primary casualties. That makes GFS’s 3.61% decline look like broad semiconductor beta, while TSEM’s 7.83% drawdown looks more like sympathy selling and industry de-rating than a direct tariff hit. Soitec belongs in the same general camp: its industrial footprint for SOI and photonics-SOI is centered on Bernin and Pasir Ris, not China or Taiwan, which again argues for a lower direct tariff multiple compression than the market appears to be applying.  EARLY-STAGE MATERIALS AND PICS The early-stage photonics material and PIC names are being traded primarily on narrative sensitivity rather than immediate tariff mathematics. POET is the clearest example. The company did report small NRE and product revenue in 2025 and disclosed an initial production-volume optical engine order in September 2025, so it is not strictly pre-revenue. But commercial scale remains immaterial relative to larger transceiver and substrate vendors. That sharply limits near-term exposure to tariff pass-through, margin compression, and factory relocation costs. Development-stage materials platforms such as Lightwave Logic are even further removed from current tariff incidence because present revenue exposure is low and valuation depends primarily on commercialization timelines, design wins, and financing conditions. When these stocks sell off alongside AAOI or AXTI, the move is mostly a multiple and risk-appetite phenomenon rather than a direct policy transmission mechanism.  Aeluma’s relative strength is understandable but should be interpreted carefully. The company states that it operates R&D and manufacturing capabilities for semiconductor wafer production, quick-turn chip fabrication, rapid prototyping, test, and validation in Goleta, California, and on March 10 it announced the appointment of a VP of Materials Operations to lead scaling of its large-diameter wafer technology into volume production. On a day when the market wanted domestic manufacturing, lower China exposure, and U.S.-anchored compound-semiconductor optionality, ALMU fit the desired profile and rose 4.10%. The relative-strength signal is credible as a narrative and positioning response. It is less reliable as proof of fundamental insulation across a full cycle because the company is still in the scale-up phase rather than established volume production.  EQUIPMENT, TEST, AND INFRASTRUCTURE The test and infrastructure names are mostly indirect risk rather than direct policy targets. Keysight’s 2025 annual report states that FY25 revenue returned to growth while managing tariff headwinds, with demand supported by wireline, semiconductor, AI-supply-chain test, and silicon photonics. Corning’s Q4 2025 release shows Optical Communications sales up 35% year over year for FY25 to $6.274 billion, indicating a still-strong optical demand environment. FormFactor remains fundamentally tied to probe cards and wafer test, while Aehr identifies silicon photonics as 1 of its served markets. None of these businesses maps closely onto the current narrow advanced-chip tariff list. Their vulnerability is instead indirect: customer order timing, semiconductor-capex duration, supply-chain friction, and a general de-rating of AI-infrastructure beta. That makes the downside in these names more a function of market structure and demand-duration fear than current Section 232 economics.  TAIWAN VERSUS CHINA The most important analytical error in the broad tape is the treatment of Taiwan and China as a single risk bucket. That is no longer accurate. Reuters reported on January 15 that the U.S. and Taiwan struck a trade deal centered on semiconductors that cuts tariffs on many Taiwanese exports and provides lower or even duty-free treatment for some semiconductor and related product imports tied to U.S. expansion. Reuters then reported on February 12 that the broader trade agreement was finalized, and on February 24 that Taiwan said the preferential tariff treatment for its semiconductor exports would not change. None of that eliminates Taiwan-linked risk. A company can still face exposure through non-preferential product categories, qualification delays, customer concentration, or later policy changes. But it materially weakens the blanket claim that Taiwan-heavy supply chains should be discounted in the same manner as China-heavy supply chains into the April 14 checkpoint. China remains the more acute source of actual export-control, permit, and retaliation risk in this group.  WHAT THE MARKET IS ACTUALLY PRICING The market is not pricing the current tariff schedule. It is pricing a compound scenario in which 1) the April 14 update signals poor negotiation progress, 2) phase 2 tariffs expand from a narrow AI-chip list into semiconductors, substrates, wafers, derivative products, or related equipment, 3) China tightens export controls further, especially around indium phosphide and other photonics-critical inputs, and 4) customers slow order placement until policy visibility improves. That scenario is clearly adverse for AAOI, AXTI, and MRVL. It is moderately adverse for LITE and COHR. It is far less directly adverse for CIEN, FN, GFS, and TSEM. The breadth of the March 30 selloff therefore suggests that the market compressed a plausible tail scenario into names with very different exposure profiles and very different capacities to localize production or benefit from domestic buildout incentives.  KEY DATES AND SIGNALS INTO APRIL 14 The highest-value signals into April 14 are not broad sector headlines but document-level details. The most important variables are whether the Commerce-USTR update describes meaningful negotiation progress with partner jurisdictions; whether any broader tariff language explicitly includes semiconductor substrates, bare wafers, optical modules, or derivative products relevant to data-center interconnect; whether the tariff-offset program is defined in a way that advantages domestic or U.S.-expanding capacity; whether Taiwan’s preferential treatment remains clearly intact; and whether China’s permit cadence for indium phosphide improves or deteriorates. A 2nd but still important date is the 180-day window implied in the proclamation, which leaves room for a later phase-2 escalation even if April 14 produces only an interim update.  BOTTOM LINE The source note captures a real policy catalyst and correctly identifies that the photonics complex sits close enough to AI semiconductor policy to trade on tariff fear. The note is most accurate at the exposed tail, especially AAOI and AXTI. It is least accurate in treating April 14 as a hard broad-tariff deadline, in collapsing Taiwan and China into a single exposure bucket, and in extending the same direct-risk logic to diversified or non-China and non-Taiwan names such as CIEN, FN, GFS, and TSEM. The current Section 232 regime is narrow, heavily carved out, and explicitly designed to preserve U.S. AI infrastructure buildout. March 30 price action therefore looks like a rational fear trade at the edges and an over-broad de-risking event across the middle. The correct interpretation is not that the entire AI photonics stack has suddenly become directly tariff-impaired. The correct interpretation is that the market has rapidly marked up the probability that localization, export controls, and a broader phase-2 tariff regime will eventually matter, and then sold nearly everything with AI-optics exposure before the legal details are known. 
CPO Landscape Mirae Analyst Note: Scale-Across: CPO ASIC: $AVGO, $MRVL Optical Transceiver: $COHR, $LITE, Innolight DSP/PAM4: $AVGO, $MRVL Coherent DCI: $CIEN, $NOK OCS Equipment: iPronics, Polatis Optical Cable / Fiber: $GLW, Prysmian, Furukawa HCF: $LITE, OFS DCI Coherent: Ciena, Nokia, Huawei Optical Amplifier: $LITE, $COHR OCS Gateway: KDDI Scale-Up: SiPh Foundry: imec, $GFS, $TSM SiPh Modulator: $NVDA (in-house MRM), $INTC ELS: NTT, Furukawa, $LITE, $COHR THz Interconnect: R&D Stage? CPO Test: "Expanding entry of new players" Micro Lens / Optical Systems: "Expanding entry of new players" TLDR: Scale Up CPO is coming next. Think the analyst note missed a bunch of upstream names and conflated ELS with light source. But it's helpful to see who they think the leading players are as a very high-level view.
Big trade (it wasn't me, as much as I would've liked it to be) in the $CIEN 1/21/28 c600 today. I feel this thing continuing to drive higher. https://t.co/pcoI8vccnc
$GLW will reclaim all time highs and continue its upward march. They are the undisputed global leader in optical fiber and control over 50% global market share in certain of the most complex/technical fiber SKUs. It is another name, like $CIEN , that institutional investors feel comfortable throwing money behind, given it was founded in 1851 (yes, a basic rationale) and people know the company, along with Wendell Weeks, very well.
$CIEN continuing to rip. Feels cheap at $60bn EV. As basic a rationale as it seems, CIEN is a name institutional investors can feel comfortable throwing money behind. The company went public in February 1997, and people know the company very well. https://t.co/2IQyjNoZpx
$CIEN EXECUTIVE OVERVIEW The OFC 2026 presentation is best interpreted as a strategic repositioning document rather than a near-term revenue bridge. Pages 3-4 and 17 map Ciena from its historical optical core into 3 adjacent AI connectivity domains: scale-across between data centers, scale-out and scale-up inside the data center, and infrastructure management through DCOM. The central message is that Ciena is attempting to evolve from a transport-centric optical vendor into a broader AI interconnect platform while still monetizing from the existing WAN and DCI franchise. That distinction matters because the current earnings engine and the future option set have very different maturity profiles and timelines.   The deck materially strengthens strategic confidence, but it does not justify treating every showcased product as a 2026 revenue driver. The existing business remains the dominant source of value creation. In Q1 FY26, revenue was $1.43B, adjusted gross margin was 44.7%, adjusted operating margin was 17.9%, and adjusted EPS was $1.35. Optical networking still represented 71.7% of revenue in Q1, while routing and switching represented 8.8%. This remains overwhelmingly an optical systems company, and the presentation should be read as an expansion of growth duration and addressable market rather than as evidence that CPO, 1.6T pluggables, or DCOM are about to dominate the P&L.   FINANCIAL AND MIX READ-THROUGH The most important validation in the deck is that the AI narrative is already visible in reported numbers rather than resting purely on prototypes. In the Q1 FY26 earnings presentation, non-telco represented 53% of revenue and direct cloud providers represented 42% of revenue, up 76% YoY. Management also disclosed that optical revenue was up more than 40% YoY, with Waveserver and RLS each up more than 80% YoY. That indicates the AI build cycle is already showing up in transport, DCI, and cloud-facing optical content, which is materially more valuable than demo-stage narrative alone.   That mix shift brings a meaningful concentration tradeoff. Ciena disclosed that 3 customers represented 47.4% of Q1 revenue, including 2 global cloud providers and 1 tier-1 North American service provider with strong MOFN activity. The earnings call also disclosed backlog of roughly $7B, up about $2B sequentially, with nearly all new orders now slated for FY27 fulfillment. This is powerful visibility, but it also means the near-term limiter is conversion capacity and customer program timing, not demand generation. The stock can therefore react to supply execution even if demand remains exceptionally strong.  The guidance framework is strong and, more importantly, plausible. Q2 FY26 revenue guidance is $1.5B ± $50M and FY26 guidance is $5.9B-$6.3B. Using the midpoint, Ciena needs roughly $1.59B of quarterly revenue in 2H26 after a $1.427B Q1 and a $1.5B Q2 midpoint. That is a real ramp, but not an extreme one given the backlog, cloud mix, and capacity build underway. The operating model also implies real leverage: gross margin is guided to 43.5%-44.5%, OpEx to roughly $1.52B-$1.53B, and management continues to frame mid-40s gross margin as a waypoint rather than an end state.   The quality of earnings is also stronger than a simple cyclical upturn would imply. The deck attributes margin improvement to capacity infill mix, engineering cost reductions, and price increases, while the earnings call indicates that pricing resets are primarily on new orders and should become more visible in Q3 and Q4. Q1 cash from operations was $228M, free cash flow was $154M, cash and investments were about $1.4B, and net debt was $174M. At the same time, CapEx of $74M was 2x-3x the average of the prior 12 quarters, reflecting capacity expansion rather than distress. This is the profile of a company investing into constraint, not a company stretching to manufacture demand.   HYPER-RAIL AND SCALE-ACROSS The most strategically significant part of the deck is Hyper-Rail. Pages 10-12 identify a very specific bottleneck in scale-across AI networks: amplifier huts on regional and long-haul routes are geographically isolated, power constrained at 3 kW per rack, limited to 4 fiber pairs per rack, and cost roughly $1M per hut to build. Ciena’s argument is that AI training and inference traffic are driving demand for 1000s of fibers across inter-data-center routes, making conventional amplifier-site architecture economically and operationally inefficient. Hyper-Rail is designed to solve precisely that constraint.   The product logic is compelling. The deck shows a 20 petabit/sec example in which today’s architecture would require 22 huts, whereas Hyper-Rail would compress that into 1 hut with 128 rails per rack, implying 32x density improvement in the same space and power envelope. Ciena’s March 10 OFC release further states that the new amplifier configurations can reduce power consumption by up to 75% and space requirements by 85%. If those economics hold in deployment, Hyper-Rail could materially increase content per route while improving customer total cost of ownership on exactly the routes that matter most for distributed AI training.   The timing, however, is clearly later than the excitement level of the slide. Management stated on the Q1 FY26 call that Hyper-Rail prototypes would be demonstrated at OFC 2026, standardization is expected to begin at the end of 2026, and ramp is expected in 2027. This is therefore a 2027+ content expansion opportunity, not a major FY26 number driver. The deck is strongest where it shows that Ciena understands the next bottleneck in scale-across AI networks; it is weaker if read as evidence of imminent revenue materiality.   The scale-across demand backdrop itself appears credible. Management disclosed that 3 hyperscalers are already using Ciena optical solutions for training applications across distance and that all 3 are ramping, including additional clusters from the earliest announced customer. The broader capex environment also supports the direction of the thesis: Meta guided 2026 capex to $115B-$135B, Alphabet guided $175B-$185B, and Microsoft reported $37.5B of FY26 Q2 capex while stating that demand continues to exceed supply. The deck’s “$600B+” aggregate 2026 hyperscaler capex number should be treated as company framing rather than a fully reconcilable public guidance figure, but the directional point is valid: AI infrastructure budgets are large enough to support a structurally larger market for inter-data-center optical connectivity.   COHERENT ROADMAP The coherent-pluggables section is nuanced, and the near-term opportunity is 800G rather than 1.6T. Page 13 focuses on WaveLogic 6 Nano 800G PKT-MAX at 135 GBaud, with both C-band and L-band variants and multi-vendor 800ZR and 800G ZR+ interoperability demonstrated at OIF. That lines up well with the standards landscape: OIF released the 800ZR Implementation Agreement in October 2024 and is demonstrating live 800ZR interoperability at OFC 2026. That makes the 800G scale-across story commercially credible and standards-backed rather than purely conceptual.   By contrast, page 14 is more roadmap than model input. The 1600ZR/1600ZR+ designs and 2nm silicon under test are important signals of technical ambition, but OIF still describes 1600ZR and 1600ZR+ as active project work aimed at defining interoperable 1600 Gbps coherent interfaces. Full-spectrum transponders are strategically interesting because they would allow operators to light an entire fiber pair’s capacity on day 1, reducing hardware overhead and deployment friction, but they remain part of a future-state architecture rather than a current shipment driver. Near-term estimates should therefore place far more weight on shipping WL6e/WL6n platforms, RLS, and Waveserver than on 1600ZR or FST. That distinction is reinforced by the Q1 FY26 earnings presentation, which disclosed that WaveLogic 6 Extreme added 18 new customers in Q1 to reach 90 total, underscoring that the monetization engine remains the current coherent franchise.   INSIDE THE DATA CENTER The inside-data-center section is strategically important but economically early. Page 15 and the February 25 press release introduce Vesta 200 6.4T CPX, described by Ciena as the highest-density, lowest-power pluggable CPO solution and designed to reduce power by up to 70% versus traditional retimed options. Nitro 2004, also from the Nubis acquisition, is a linear redriver aimed at extending copper reach for scale-up networks and is claimed to reduce power by up to 80% versus AEC solutions. Both products are sampling in Q2 CY26. That timing suggests FY26 contribution is likely to be dominated by design engagement, qualification, and ecosystem positioning rather than material revenue.   The Nubis acquisition should therefore be viewed as long-duration strategic positioning rather than near-term financial acceleration. Ciena’s acquisition presentation stated that Nubis was expected to have no material impact on revenue or operating expense in FY25 or FY26 and to generate positive profit contribution in FY28. That is a critical anchor against over-extrapolating the OFC deck. The more appropriate interpretation is that Ciena is deliberately building an architecture-agnostic interconnect portfolio spanning coherent pluggables, CPO, co-packaged copper compatibility, linear copper extension, and future coherent-lite building blocks. In a market where the long-run split among copper, IMDD, CPO, and coherent inside the data center is still evolving, that breadth is strategically attractive even if the P&L impact is deferred.   DCOM is the most underappreciated adjacency in the deck because it is less visually dramatic than optical demos but closer to commercial proof. The presentation states that DCOM was created via Meta collaboration, is designed for hyperscale provisioning and configuration requirements, and is in technical discussions with 2 additional major hyperscalers. Ciena’s DCOM materials state that the PON-based architecture can cut rack space by 99% while reducing power and operational complexity, and the Q4 FY25 earnings presentation disclosed that the solution is already operational in Meta data centers. That does not make DCOM large today, but it does distinguish it from purely future-stage products. It is a control-plane and infrastructure-management wedge that could create sticky software plus hardware content in a pain point that scales with AI cluster size.   STRATEGIC AND COMPETITIVE IMPLICATIONS The deepest strategic advantage visible in the deck is not any 1 component claim; it is portfolio coherence. Ciena can connect the inter-data-center backbone via RLS, Waveserver, and coherent pluggables; attack amplifier-site economics via Hyper-Rail; participate in inside-DC interconnect via Vesta and Nitro; and simplify the management plane via DCOM. That is unusual breadth for a company whose revenue base is still dominated by WAN optical transport. The speaking agenda reinforces technical relevance by placing Ciena executives and engineers alongside Meta, NVIDIA, Arista, Broadcom, Marvell, Google, and others in discussions on scale-up, pluggables, CPO, and 1.6T+. That is not proof of share, but it is evidence that Ciena is present in the correct architectural conversations.   The counterpoint is equally important. Commercial maturity is highly uneven across the portfolio. RLS, Waveserver, WL6e, and 800G pluggables are real businesses with customer traction, revenue scale, and backlog support. Hyper-Rail is pre-standardization. 1600ZR and FST are roadmap constructs. Vesta and Nitro are at sample stage. DCOM appears commercially real but remains small relative to the corporate revenue base. The deck therefore broadens terminal TAM more clearly than it changes the 2026 P&L. That is the critical distinction for valuation work and estimate discipline.   RISKS AND MONITORING POINTS 4 risks stand out. 1st, supply remains the gating factor. Management stated that Q1 revenue would have been higher absent constraints and that demand is expected to outstrip supply for at least the next several quarters. 2nd, customer concentration is high and increasingly cloud-centric, which raises exposure to program timing and capex digestion. 3rd, several of the deck’s most exciting technologies are not yet standardized or broadly deployed. 4th, inside-DC monetization depends on architecture choices that remain unsettled across copper, IMDD, CPO, and coherent. None of these 4 risks invalidates the thesis, but all 4 limit how much future-stage optionality should be capitalized in near-term forecasts.   The monitoring framework is straightforward. In the current business, the key questions are whether RLS, Waveserver, and pluggables continue to convert backlog into shipments, whether gross margin keeps trending toward the mid-40s, and whether capacity additions relieve supply bottlenecks without causing working-capital slippage. In the option set, the key questions are whether Hyper-Rail moves from prototype to standardization on schedule, whether Vesta and Nitro convert sampling into named design wins, and whether DCOM expands beyond Meta into broader hyperscaler adoption. Improvement on the 1st set changes near-term estimates; improvement on the 2nd set changes growth duration and strategic multiple.   CONCLUSION The bottom line is that the OFC 2026 presentation materially improves confidence in the breadth and internal logic of Ciena’s AI networking strategy, but the monetization timelines across that strategy are uneven. The validated earnings engine is the existing optical franchise, where cloud mix, RLS, Waveserver, pluggables, backlog, and operating leverage are already visible. The deck’s longer-duration value lies in showing that Ciena has credible claims on the next layers of AI connectivity: scale-across densification through Hyper-Rail, standards-based coherent evolution toward 1.6T, and inside-data-center expansion through Vesta, Nitro, and DCOM. The appropriate analytical stance is constructive on the core, interested but selective on the options, and disciplined about not pulling 2027+ architecture upside too aggressively into 2026 numbers.  
$CIEN stacking another 52w high. Few can do what they can and especially at their scale. Would love to see them take out all time highs from back in the internet boom days. https://t.co/tjYl4x9pOx
. @aleabitoreddit is locked into this GAI infrastructure build-out cycle. I am trying to keep up. $LITE $COHR $CIEN $AAOI EXECUTIVE OVERVIEW The post advances a real but overcompressed thesis. AI cluster scaling is forcing the optical stack to move from copper-dominated and pluggable-optics-centric architectures toward deeper silicon photonics, external laser source, advanced packaging, and eventually co-packaged optics. Broadcom has already described industry 1st volume shipments of CPO with multiple partners shipping TH5-Bailly systems, and NVIDIA has formally launched photonics switch platforms with an ecosystem that includes TSMC, Coherent, Corning, Foxconn, Lumentum, and Sumitomo. The core directional claim that the bottleneck is migrating upstream into SiPh wafers, CW DFB laser arrays, fiber attach, packaging, and test is therefore credible. The weaker parts of the post are the timing and the basket construction. Commercialization is underway, but LightCounting still characterized high-volume CPO deployments as more likely in 2027, while 800G+ retimed pluggables continue to grow and are not being displaced in 1 step. (NVIDIA Investor Relations) The investment implication is that the post should not be read as a 1-cycle rotation trade from “old” optics names into “new” CPO names. It is better read through a 3-layer framework. 1st, near-term revenue remains concentrated in pluggable transceivers, optical modules, and the component set feeding 800G and 1.6T deployments. 2nd, the next architectural constraint set includes external laser sources, photonics substrates, foundry capacity, fiber attach, connector and cage systems, and assembly and test. 3rd, a large share of the highest-conviction architectural innovation remains private, which means the public basket often offers only 2nd-derivative exposure. The post is strongest when it highlights real supply-chain choke points and weakest when it assumes that technical relevance will automatically convert into material public-market earnings torque. (LightCounting) HYPOTHESIS TEST Interpreted as a claim that 2026 is the start of meaningful commercialization, the timing is defensible. Interpreted as a claim that 2026 is the year of broad mass deployment, it is aggressive. NVIDIA guided Quantum-X Photonics InfiniBand switches for availability later in 2025 and Spectrum-X Photonics Ethernet switches for 2026. Broadcom states that multiple partners are already shipping TH5-Bailly systems and that 200G/lane Gen 3 CPO is in volume shipment. That is enough to confirm that CPO is no longer purely experimental. However, LightCounting’s 2025 conference summary still described high-volume deployments as likely in 2027 and explicitly stated that retimed pluggables are not going away, with 800G and higher-speed transceivers expected to triple from 2025 to 2030. The post therefore captures the direction of travel better than the cadence of revenue realization. (NVIDIA Newsroom) The post is also correct that value capture should rotate toward CW DFB and SiPh architectures as the industry moves from EML-heavy pluggable designs toward external laser source and integrated photonics engines. But the rotation is not cleanly from 1 set of companies to a different set. Lumentum and Coherent, the names presented as “current EML bottlenecks,” are also active participants in external laser source modules for CPO. Broadcom is not simply a light-source supplier but the current merchant platform owner in CPO. Marvell is not merely a mixed-signal supplier but a leading 1.6T DSP and light-engine platform company. The post’s basket is therefore directionally insightful but analytically underweighted toward the true platform owners and overweighted toward smaller, higher-beta derivative names. (Lumentum) LIGHT SOURCE AND LASER SUPPLIERS Sivers Semiconductors (SIVE) is 1 of the cleaner pure-play expressions of the CW DFB laser-array thesis. The company is explicitly marketing high-power CW DFB lasers for AI optical interconnects, silicon photonics, optical I/O, and CPO. It expanded its partnership with Ayar Labs around product qualification and manufacturing readiness for high-volume optical I/O deployments, and it partnered with POET on external light sources for future pluggable and CPO solutions. The post is directionally correct to frame Sivers as a “next cycle” name, but it remains a small-cap, pre-scale exposure where qualification wins and manufacturing ramp timing matter more than architecture alone. (Sivers Semiconductors) Sumitomo Electric is a credible incumbent rather than a speculative bottleneck trade. Management outlined 100mW and 350mW CW-LD products, explicit support for high-density SiPh modulators for 1.6T OSFP and CPO, and a 4x increase in LD chip production capacity. The post is correct that Sumitomo belongs in the CW DFB ecosystem, but the equity expression is less pure because CPO is only 1 part of a large diversified industrial portfolio. The technological fit is strong; the public-market sensitivity is more muted. (Sumitomo Electric) Lumentum (LITE) is mischaracterized if treated as only a current-cycle EML bottleneck. Its ELSFP module is explicitly designed as a centralized, serviceable light source for CPO systems, allowing multiple SiPh optical engines to share a single ultra-high-power laser source. Lumentum therefore straddles 2 cycles: it benefits from current pluggables and has a real seat in the external-laser-source architecture that the post frames as the next bottleneck. The correct debate is not whether Lumentum is “old” or “new,” but how much of its future mix migrates from EML content toward CPO laser modules and adjacent photonics. (Lumentum) Coherent (COHR) is similarly more than an EML carry trade. Its ELSFP-compliant module integrates 8 high-power 1310 nm lasers for CPO and next-generation optical systems, and the company is also demonstrating a silicon-photonics-based 1.6T-DR8 transceiver module using Marvell’s Ara DSP. In practice, Coherent participates across DCO, pluggables, lasers, silicon photonics modules, and CPO-enabling components. The post is correct that Coherent is central to today’s optical bottlenecks, but incorrect if it assumes those bottlenecks disappear from Coherent’s addressable set in the next architecture. (Coherent Inc) Broadcom (AVGO) is the most important name in the entire post and is understated by being listed merely under “light source.” Broadcom states that multiple partners are now shipping TH5-Bailly systems, describes these as the industry’s 1st volume shipments of CPO, and has already advanced to 200G/lane Gen 3 CPO with Gen 4 underway. Broadcom is not simply supplying an optical component. It is defining the merchant switch, SerDes, silicon photonics, ecosystem, and CPO system architecture. Any hedge-fund-quality reading of the post should elevate Broadcom from a component bucket to the central platform owner in listed CPO. (Broadcom Documentation) MACOM (MTSI) is a credible enabler, but the post’s “light source” and “analog/mixed-signal” labels both understate the complexity of its role. MACOM launched 4 new 200G/lane solutions for 1.6T optical connectivity and is highlighting a portfolio that includes photodetectors, TIAs, laser drivers, and receive-side optical chip stacks. The company is relevant wherever 200G/lane optics are deployed, whether in advanced pluggables or early CPO designs. The right characterization is “critical component supplier” rather than “pure CW DFB winner.” (Macom) Applied Optoelectronics (AAOI) is accurately framed by the post as a way to ride the current pluggable bottleneck. AOI has been ramping 400G and 800G transceiver capacity with a roadmap to 1.6T, has announced an ultra-high-power semiconductor laser for silicon photonics and CPO, and in 2026 disclosed its 1st volume order for 1.6T data-center transceivers from a major hyperscale customer. The key point is that AAOI is a nearer-term transceiver and current-cycle optics beneficiary with optionality on future CPO laser content, not yet a proven pure-play CPO bottleneck owner. (AOI Newsroom) SIPH FOUNDRIES Tower Semiconductor (TSEM) is a legitimate open-foundry SiPh exposure. Tower’s PH18 SiPho offering targets the o-band and c-band data-center interconnect market through its 200 mm open foundry platform and MPW shuttle structure. The investment case is ecosystem participation and design-in leverage rather than sole-volume leadership. The post is directionally correct to include Tower, especially for investors seeking foundry participation without relying exclusively on captive internal platforms. (Tower Semiconductor) GlobalFoundries (GFS) is 1 of the more credible listed foundry expressions of the SiPh thesis. GF’s Fotonix platform is explicitly positioned for ultra-fast, ultra-efficient silicon photonics, and GF disclosed that Ayar Labs’ UCIe optical interconnect chiplet will utilize GF’s monolithic photonics platform. That matters because it links GF to 1 of the most credible private optical I/O innovators. The post is correct to include GFS, and the strategic value is arguably higher than its treatment in many retail CPO baskets. (GlobalFoundries) UMC is the weakest and earliest-stage foundry inclusion in the post. The company only recently disclosed that licensed technology from imec will enable UMC to bring a 12-inch silicon-photonics platform to market targeting next-generation connectivity. That is real progress, but it is still better described as a platform-formation story than an earnings-visible SiPh ramp. The post is not wrong to include UMC, but the implied proximity to monetization is ahead of the evidence. (UMC) TSMC (TSM) is a highly credible but financially diluted CPO exposure. TSMC has described an advanced silicon-photonics foundry platform for next-generation data communications, and NVIDIA’s photonics ecosystem explicitly includes TSMC. This makes TSMC a foundational manufacturing beneficiary of any large-scale CPO rollout, especially where advanced logic, SiPh, and 3D integration must be co-optimized. The limitation is not strategic relevance but public-market purity: even successful CPO ramps are small relative to TSMC’s total earnings base in the near term. (TSMC Research) Intel (INTC) remains technically credible in silicon photonics despite broader corporate complexity. Intel has presented a fully integrated optical I/O chiplet aimed at co-packaged optical I/O for AI infrastructure and describes a path to detachable optical connectors and co-packaging with next-generation CPU, GPU, IPU, and SoC products. The post is correct that Intel belongs in the SiPh foundry and integration conversation. The challenge is that optical progress must be weighed against the company’s much larger foundry, CPU, and restructuring narratives. (Newsroom) DCO AND COHERENT PLUGGABLES This section is the largest conceptual mismatch in the post. DCO demand is real and AI-driven, but it primarily addresses routed optical, metro, and data-center-interconnect bandwidth rather than the in-package optical bottlenecks that define CPO. These are valid AI optical-networking exposures, not pure CPO bottleneck exposures. (Nokia Corporation | Nokia) Nokia (NOK) is a coherent-pluggable and routed-optical beneficiary, not a direct CPO bottleneck owner. Its ICE-X 800G ZR/ZR+ platform is clearly relevant to AI-scale connectivity and long-haul-capable 800G transport, but that sits primarily in inter-data-center and transport networks rather than in-package or in-rack CPO. The post is correct only if “CPO supply chain” is being used loosely to describe the broader optical-networking stack created by AI traffic growth. As a direct CPO expression, Nokia is materially less pure than the post implies. (Nokia Corporation | Nokia) Ciena (CIEN) is similarly a coherent and DCI beneficiary rather than a direct CPO bottleneck. WaveLogic 6 spans 800G coherent pluggables and the industry’s 1st 1.6 Tb/s transceivers, and Ciena has highlighted 800G coherent pluggables as an important part of hyperscaler and cloud ramp activity. This makes Ciena relevant to AI transport and multi-site connectivity. It does not make Ciena a primary beneficiary of co-packaged optics inside AI back-end switches. The post blurs 2 different optical problems by grouping them together. (Ciena) Cisco (CSCO) belongs in the same bucket. Cisco guided its 1st coherent pluggable optics for spring 2025 and continues to position optical networking and Silicon One routing as foundational for distributed AI workloads. That is meaningful for scale-out and DCI architecture. It is not the same as being a direct supplier into SiPh CPO packaging inside merchant AI switches. The post again conflates routed optical networking with co-packaged optics. (Cisco Newsroom) Coherent (COHR), in its DCO role, is more credible than the other 3 because it participates across both coherent pluggables and the CPO-adjacent hardware stack. Its 800G ZR/ZR+ pluggables serve metro, regional, and DCI needs, while the company simultaneously develops 1.6T silicon photonics modules and CPO laser sources. That cross-stack presence makes Coherent a better optical-infrastructure basket name than a pure DCO proxy. (Coherent Inc) MICRO-LENS AND FIBER ARRAY SUPPLIERS Himax (HIMX) is 1 of the most speculative names in the post. The company’s relevance comes through proprietary wafer-level optics and nano-imprinting capabilities plus its strategic relationship with FOCI, not from a directly disclosed hyperscale CPO revenue stream. Himax has stated that its CPO collaboration with FOCI is ongoing, that 1st-generation solutions are being validated by key customers, and that the companies are working toward mass-production readiness in 2026. That makes the exposure real, but still early and highly contingent on external validation and scale conversion. (Himax) FOCI (3363.TWO) is much closer to a true micro-bottleneck supplier. Its ReLFACon connector is explicitly designed to install fiber-array connectors into silicon-photonics CPO components, support 8 to 128 channels, meet standard semiconductor-packaging environments, and achieve mass automated production. If fiber attach and optical coupling become a hard manufacturing bottleneck in CPO, FOCI’s relevance could be material. The constraint is not technical fit but visibility into customer concentration, scale, and public-market liquidity. (Foci) OPTICAL INTERPOSERS POET Technologies (POET) is correctly categorized as an optical interposer name, and the post’s inclusion is logically sound. POET’s Optical Interposer platform is designed to unite photonic and electronic devices in a single chip-scale package, the company has demonstrated 1.6T optical engines and light-source products, and it has disclosed both a production order for 800G optical engines and an external-light-source collaboration with Sivers for CPO. The caution is that POET remains a platform-validation story rather than a proven volume leader. The technology is relevant; the earnings power remains execution dependent. (POET Technologies) SUBSTRATES Soitec (SOI) is 1 of the strongest calls in the entire post. Management explicitly identifies Soitec as a leading supplier in Photonics-SOI for optical transceivers, states that silicon photonics is the leading CPO technology selected by industry players such as NVIDIA, Broadcom, and Marvell, and in recent reporting described CPO components as reliant on specialist Photonics-SOI substrates where Soitec is a leader. This is exactly the kind of upstream bottleneck where content can rise before the market fully appreciates the architectural transition. Of the 3 names the post highlights as “perfect examples,” Soitec is the most credible. (Default) AXT (AXTI) is a more complicated substrate call. The company is benefiting from rising indium-phosphide demand tied to AI infrastructure and high-speed optical connectivity, and management says it is on track to more than double InP capacity. However, the same company has also disclosed that quarterly results were affected by fewer export-control permits for indium phosphide issued by China’s Ministry of Commerce. The post is correct that AXT sits near an important material bottleneck, but the bottleneck is not only technical. It is also geopolitical and regulatory. That makes AXT a high-beta substrate expression rather than a clean 1. (AXT Investor Relations) Shin-Etsu is a credible but diluted substrate exposure. The company is a major SOI-wafer producer and management has explicitly cited silicon photonics as part of diversification into new applications. The post is directionally correct to include it, but the public-market expression is much broader than CPO and lacks the concentrated torque of Soitec’s Photonics-SOI franchise. Shin-Etsu should be treated as a diversified wafer incumbent with silicon-photonics optionality, not as a pure-play next-bottleneck name. (Shinetsu) OPTICAL PACKAGING AND ASSEMBLY Fabrinet (FN) is a strong near-term beneficiary of optical complexity but not necessarily a unique CPO bottleneck owner. The company remains heavily exposed to optical communications, and recent results showed strong data-center-interconnect growth. The logic of including Fabrinet is that increasing optical-module complexity and AI-related volume growth lift outsourced packaging and manufacturing intensity. The limitation is that Fabrinet captures value as an execution-heavy manufacturing partner rather than as the owner of a unique architectural chokepoint. (Fabrinet Investor Relations) ASE Technology Holding (ASX) is a genuine strategic CPO enabler. ASE introduced a silicon-photonics packaging platform, publicly discussed the rise of SiPh and CPO as a driver of OSAT demand, and in 2025 demonstrated a CPO device that solves the assembly challenge of multiple optical engines with an ASIC in a large integrated package. The post is correct to include ASX. The main nuance is that CPO is 1 growth vector inside a much larger advanced-packaging business already driven by AI, HBM, and chiplets. (ASE) Innolight is better viewed as a current-cycle module winner than as a hidden next-cycle bottleneck. The company states that it has invested heavily in silicon photonics and other key platforms to support 100G through 1.6T transceivers. That makes it very relevant to immediate hyperscale optical spend. It does not automatically make it the place where the incremental economic bottleneck of CPO will sit. The post includes Innolight for a good reason, but the revenue timing is more current-cycle than next-cycle. (InnoLight Technology) Eoptolink fits the same pattern. The company has been demonstrating 2nd-generation fully retimed 1.6T OSFP transceivers and broader 1.6T, 800G, LPO, and LRO solutions. It is also signaling work on next-generation 400G/lane products. This is highly relevant to current 1.6T module ramps and to future optical roadmaps, but it remains more of a transceiver vendor than a bottleneck owner in the strict architectural sense used by the post. (Eoptolink) ANALOG AND MIXED-SIGNAL ICS MACOM (MTSI) is a real beneficiary of 200G/lane adoption across pluggables and early CPO. Its 4 new 200G/lane solutions for 1.6T optical connectivity and related laser driver, TIA, and receive-side products make it relevant wherever the industry standardizes around 200G/lane optics. The post is right to include MACOM, but the expected earnings sensitivity depends on socket share against larger or more integrated peers. (Macom) Semtech (SMTC) is a credible signal-integrity beneficiary. The company introduced TIAs for the emerging 1.6T optical-interconnect market, claims up to 20% power reduction for certain 800G deployments, and framed its 200G/channel PMD portfolio as the next wave of building blocks for 1.6T modules, with 400G/channel development underway for 3.2T. The post is correct that Semtech belongs in the supporting mixed-signal layer. The company is best viewed as a component supplier leveraged to optical bandwidth density rather than as a central CPO platform owner. (Semtech) Marvell (MRVL) is materially more important than the post’s “analog/mixed-signal” label suggests. Marvell states that Ara is now shipping in mass volume to global customers, has demonstrated a 1.6T silicon-photonics light engine as a foundation for CPO system development, and is showcasing CPO and CPC system-level platforms for scale-up and scale-out fabrics. Marvell therefore spans current 1.6T monetization and future CPO integration. In listed markets, Marvell is 1 of the highest-quality ways to own the transition from pluggables toward deeper optical integration. (Marvell Technology) MaxLinear (MXL) is a legitimate 1.6T interconnect name, but still a challenger. Rushmore is sampling and commercially available as a low-power 1.6T 8x200G PAM4 DSP enabling 1.6T optics and active copper cables, with a product page describing DR8 and 2xFR4 reach profiles. The relevance is real. The remaining question is scale, customer-adoption breadth, and whether MaxLinear can capture enough volume to matter relative to Marvell and other entrenched optical-DSP suppliers. (MaxLinear) SPECULATIVE MODULATOR MATERIALS Lightwave Logic (LWLG) is accurately labeled speculative. The company’s core claim is that EO polymers can transmit more data at higher speeds with lower power, and public disclosures point to a development agreement with Tower for modulators on the PH18 silicon-photonics platform. That keeps LWLG within the realm of genuine technology relevance. It does not yet establish broad commercial adoption, material customer volume, or durable economic ownership of the next CPO bottleneck. The post is directionally fair on the science and too aggressive on investability. (Lightwave Logic) CONNECTORS AND FIBERS Corning (GLW) is 1 of the most credible enabling names in the post. Corning is supplying optical components for Broadcom’s Bailly CPO system and launched CPO FlexConnect fiber specifically optimized for short-length inside-the-box CPO deployments. Corning is therefore directly embedded in the merchant CPO ecosystem, not merely a generic fiber beneficiary. The limitation is diversification, but the strategic fit is strong. (Corning) Amphenol (APH) is a broad AI interconnect beneficiary rather than a narrow CPO bottleneck expression. The company is pushing 224G connector and cage systems, high-speed cables, and XPO optics-over-copper solutions for next-generation data centers. That makes it well positioned for the broader AI interconnect build-out. The post’s inclusion is reasonable, but the exposure is more “general high-speed interconnect density” than “specific CPO chokepoint.” (Amphenol) TE Connectivity (TEL) has a more explicit CPO claim than Amphenol. TE’s ELSFP connector and cage is specifically designed for 102.4T CPO systems under the OIF ELSFP implementation agreement. That makes TE a direct enabler of the external-laser-source architecture that sits at the center of next-generation CPO serviceability. The post is correct to include TEL and arguably understates how direct that inclusion is. (TE Connectivity) Foxconn Interconnect Technology (FIT) is an underappreciated name in the list. FIT disclosed that it is delivering solderless LGA-to-LGA sockets, remote pluggable laser-source cages, and specialized connector assemblies that are integral to Broadcom’s TH5-Bailly platform. Because CPO requires not only photonics but serviceable mechanical and interconnect architecture, FIT is closer to a real bottleneck beneficiary than the market’s conventional categorization would suggest. The post is correct here. (Fit Foxconn) Fujikura is a legitimate fiber-infrastructure beneficiary, but it is more levered to dense cabling and backbone-fiber scale-out than to in-package CPO economics. The company has been launching ultra-high-fiber-count cable products and discussing the role of backbone fiber in AI networks. That is strategically relevant as AI clusters distribute across facilities and regions. It is less direct to the specific CPO bottlenecks described in the post. (Fujikura)
$CIEN taking out 52w high today. Post-print sell-off was off base. I see this name continuing to move higher from here. https://t.co/2ggZ1EgVw1
$CIEN is continuing to recover from its post-print gap down. Getting some big volume bars today. https://t.co/OYPnyOPSuB
$CIEN Ciena Corp (CIEN) Q1 FY2026 Post-Earnings Debrief https://t.co/73ivqmT3LS CIEN sold off -14.4% over 2 days despite delivering the best quarter in the company's history. Revenue of $1.43B beat consensus by 10.4%, adjusted EPS of $1.35 beat by 69.8%, EBITDA beat by 29.6%, and gross margin beat by 128 bps. Operating margin inflected to 13.3% from 0.8% in Q4. The selloff was emphatically NOT about Q1 results.
Very few companies can do what $CIEN can do, and hardly at their scale. (I might be a little bit more quite, but I am still investing) https://t.co/FnESkjsfvu
$CIEN EXECUTIVE CALL SUMMARY: (Ciena Corp, 03/05/26) Fiscal Q1 2026 results were reported as record-setting on revenue and materially stronger on adjusted profitability, alongside an unusually large step-up in backlog and a raised FY26 outlook. Revenue of $1.43 billion (+33% y/y) was described as the highest quarter in company history and at the top end of prior guidance, with adjusted EPS of $1.35 (vs $0.64 y/y) and adjusted gross margin of 44.7% (ahead of expectations). Management framed demand as “incredibly strong” and “unprecedented,” citing exceptional order activity and long-term customer planning discussions as reinforcing multi-year durability. The most consequential disclosure on the call was backlog expansion of approximately $2 billion in the quarter to approximately $7 billion exiting Q1, with management stating “nearly all new orders we are taking now will be for fulfillment in fiscal 2027,” implying both unusually strong demand visibility and binding supply constraints. FY26 revenue guidance was raised to $5.9 billion to $6.3 billion, with midpoint y/y growth raised from 24% to 28%, and FY26 adjusted gross margin guidance raised to 43.5% to 44.5% (stated as 1 point above the December guide). Despite the operational beat and raised outlook, the Bloomberg note included in the source material indicated shares were down 3.5% premarket, with shares up nearly 50% YTD as of the prior close, consistent with a setup where expectations and positioning were elevated and incremental upside likely needed to exceed a high bar. Management’s strategic narrative remained tightly anchored to AI-driven connectivity spend and the “in and around the data center” expansion thesis, with particular emphasis on “scale across” distributed AI training clusters (3 hyperscalers cited as deploying Ciena optical solutions for training across distance), accelerating RLS and Waveserver demand, and an expanded interconnect roadmap (800 ZR and new CPO-related products from the Nubis acquisition). The call also highlighted MOFN as a meaningful structural driver in the service provider segment (10% to 15% of service provider business), with India called out as up 40% y/y in orders. The principal near-term limiter remains supply availability, with management explicitly stating Q1 revenue “would have been higher but for these constraints” and expecting demand to outstrip supply “at least for the next several quarters,” a dynamic that supports backlog growth and pricing/value-exchange leverage but caps near-term revenue conversion and introduces component-cost and execution risks. Q1 2026 PERFORMANCE AND MIX Reported Q1 performance showed broad strength concentrated in networking platforms and optical. Total revenue of $1.43 billion was +33% y/y, with networking platforms revenue of $1.15 billion (+40% y/y) and converged packet optical revenue of $1.02 billion (+41% y/y). Routing and switching revenue was $126.0 million (+35% y/y). Platform software and services revenue was $93.3 million (-1.9% y/y), with Blue Planet automation software and services revenue of $20.4 million (-22% y/y). Global services revenue was $164.1 million (+26% y/y), but below the $178 million estimate cited in the Bloomberg excerpt; within services, maintenance/support/training was $87.6 million (+17% y/y), installation/deployment was $67.9 million (+42% y/y), and consulting/network design was $8.6 million (+12% y/y), with installation/deployment also below the $72.7 million estimate. The mix and segment detail reinforce 2 key points from management’s narrative. First, the optical transport core is accelerating, with management citing optical revenue up “over 40%” y/y and identifying Waveserver and RLS as up “over 80%” y/y, consistent with hyperscaler WAN/DCI demand and the “scale across” training cluster buildouts. Second, software/automation performance was weak relative to the hardware acceleration, with Blue Planet down 22% y/y and consulting/network design below the estimate, suggesting the current cycle is hardware-capacity led rather than software-led. This does not disprove the longer-term “automation and services” thesis, but it does imply that near-term operating leverage is coming primarily from product scale, product mix, and cost-down actions rather than from accelerating software attach. Adjusted gross margin of 44.7% was a central upside element, especially given ongoing supply constraints. Management attributed outperformance to product mix including “incremental demand for capacity infills,” execution of cost reductions, and “early progress on advancing the value exchange with our customers.” The “capacity infill” comment is important because it implies near-term demand is not limited to greenfield buildouts; customers are also pulling incremental line-side and capacity augmentation where networks are already deployed, which typically carries shorter sales cycles, faster delivery urgency, and potentially better mix. However, reliance on infill mix can also add quarter-to-quarter variability and can reverse if delivery timing shifts due to component constraints. DEMAND SIGNALS, BACKLOG QUALITY, AND REVENUE CONVERSION The order and backlog discussion dominated the incremental information set. Backlog was stated to have increased by approximately $2 billion in the quarter, exiting Q1 at approximately $7 billion, a step-change that is difficult to attribute to a single program and was described as broad-based across service providers, hyperscalers, neoscalers, submarine, and MOFN. Management explicitly framed backlog quality as high and increasingly visible, referencing deep collaborative relationships and installation activity. Installation services were up 42% y/y, and management stated this provides “unique visibility into what they’re doing and deploying,” a direct rebuttal to the common optical-cycle risk of “phantom backlog” created by early ordering during constraints. Several specific quantitative markers help triangulate backlog composition and near-term conversion. Management stated backlog is roughly 80% products and software and the rest services. Remaining performance obligations (RPO) were indicated at roughly 60% of orders taken in Q1, implying a meaningful portion of Q1 order activity converted quickly (consistent with “capacity infills”) while a larger portion pushed into future delivery windows. The most consequential operational implication was the statement that “nearly all new orders we are taking now will be for fulfillment in fiscal 2027,” signaling lead times and/or capacity reservations that extend well beyond the current fiscal year. This combination typically indicates demand is currently supply-limited, and the revenue guidance range is framed as balancing demand with supply. The demand narrative was reinforced by several discrete end-market drivers discussed on the call. Service providers were described as having underinvested in optical infrastructure for approximately 5 years due to 5G prioritization, now reinvesting alongside “autonomous networking capabilities” to handle AI-driven traffic and improve operating efficiency. MOFN was positioned as a structural driver as cloud providers and service providers navigate regulatory and capacity requirements, with India singled out as a high-demand geography and management expecting MOFN revenue to remain an important contributor to service provider growth. Hyperscaler demand was framed as multi-wave, spanning traditional WAN/DCI, subsea, and new “scale across” training clusters, with inference highlighted as an additional growth vector “still in its infancy.” The principal counterweight to these signals is the inherent cyclicality of optical transport procurement and the historical pattern where tight supply conditions can pull forward orders. This risk was raised explicitly in Q&A. Management’s mitigation framing centered on improved end-deployment visibility (installation services, deeper collaboration), and tighter terms and conditions as part of “value exchange,” with emphasis on avoiding being “stuck holding the bag.” This reduces but does not eliminate the risk of future schedule changes, deferrals, or cancellations, particularly if hyperscaler architecture decisions shift, if service provider funding environments tighten, or if supply normalizes rapidly and customers rationalize buffered orders. SUPPLY CONSTRAINTS AND CAPACITY INVESTMENT Supply constraints were described as the binding limiter on near-term revenue, with management stating “our revenue in the first quarter would have been higher but for these constraints” and expecting demand to outstrip supply “at least for the next several quarters.” Constraint location was indicated as more acute at the component level than at final assembly, motivating a 2-pronged strategy: expanding contract manufacturer output capacity and securing/expanding component supply via long-term purchase commitments. This has 3 investment-relevant implications. Near-term upside is more a function of supply unlocks than demand creation. With demand described as exceptionally strong and backlog now extending into FY27 fulfillment, quarter-to-quarter revenue and gross margin outcomes are likely to be governed by component availability, allocation, and delivery timing. This makes the operational risk profile more execution/supply-chain driven than demand driven in the next several quarters. The capital intensity and balance-sheet posture are shifting to support growth. Q1 CapEx was $74 million, described as approximately 2x to 3x the average over the last 12 quarters, reflecting accelerated capacity investments. Management also discussed investing through FY26 with the intent to realize capacity benefits in 2027, implying elevated CapEx and potentially higher working capital requirements during the build. Q1 cash from operations was $228 million, inventory turns improved to 3.2x, and cash conversion improved by 3 days, suggesting working capital was managed well in Q1 despite rising demand. Sustaining this balance as supply commitments increase will be a key determinant of free cash flow durability. Supply tightness supports pricing and broader “value exchange” leverage, but also heightens component inflation risk. Management acknowledged cost increases from the supply chain and the possibility of vendors repricing backlogs. The stated approach was to lock in pricing with suppliers where possible, and to engage both suppliers and customers so margin is not compressed “in the middle.” This creates a dynamic where gross margin performance becomes a function of (a) pass-through or pricing leverage, (b) cost-down engineering actions, and (c) mix and delivery efficiency, rather than only volume leverage. MARGINS, PRICING, AND OPERATING LEVERAGE Adjusted gross margin printed at 44.7%, and management reiterated a milestone target of 45% gross margin, describing progress as driven by mix, cost reductions, and early “value exchange” improvements. In Q&A, management provided additional color that gross margin tailwinds came from higher customer “infill rates” (capacity increases across hyperscalers and service providers) and engineering-driven cost reductions “separate from the supply chain activities.” Importantly, pricing actions discussed previously were framed as not yet fully reflected: “the price increases that we talked about at the end of last year… haven’t started to fully kick in until the second half of the year,” with expectations that pricing will “show up in Q3 and Q4” because increases were primarily on new orders and the backlog was large. This implies 2H FY26 margin could have incremental support from pricing, but management also guided to roughly equivalent 1H and 2H gross margins, indicating either conservatism or offsetting headwinds (mix, component costs, expedite costs, or less favorable product mix as volumes scale). The decision to guide to roughly flat 1H vs 2H gross margin is a critical interpretive point because it conflicts with the intuitive expectation of 2H expansion from price increases and an 800 ZR ramp. When challenged on why 2H gross margin would not expand, management emphasized “a responsible guide” in light of product mix and supply chain challenges. This suggests either (a) meaningful uncertainty on component cost escalation and allocation, (b) mix volatility as supply-constrained deliveries skew toward lower-margin configurations, (c) deliberate customer/market-share tradeoffs limiting aggressive pricing, or (d) a desire to avoid overpromising margin expansion given the stock’s elevated expectations. Operating leverage was highlighted as a priority, with adjusted operating margin of 17.9% stated as 190 bp above the midpoint of the December guide. Management described holding OpEx approximately flat vs 2025 while increasing R&D investment intensity in high-growth segments, enabled by (i) resetting prior-year incremental OpEx guidance, (ii) a workforce reduction of 4% to 5% (savings reinvested), and (iii) ceasing further investment in a 25-gig PON initiative. FY26 adjusted OpEx was guided to $1.52 billion to $1.53 billion, with a stated “small difference” due to stronger demand. This indicates a posture of disciplined cost control while ensuring capacity and product velocity investments are funded, a combination that can sustain elevated operating margins if gross margin remains stable and revenue continues to scale. GUIDANCE, RAISES, AND WHAT CHANGED FY26 guidance was raised on both revenue and gross margin. Revenue guidance was raised to $5.9 billion to $6.3 billion, with midpoint growth raised from 24% to 28% y/y. The raise was explicitly framed as balancing demand strength with supply constraints, which implies the company is not assuming supply becomes fully unconstrained in FY26. FY26 adjusted gross margin guidance was raised to 43.5% to 44.5%, stated as 1 point above the December guide and 130 bp above FY25, while also indicating that 1H and 2H gross margins are now expected to be roughly equivalent. FY26 adjusted operating margin was guided to 17.5% to 19.5%, with OpEx guided at $1.52 billion to $1.53 billion. Q2 FY26 guidance was provided at revenue of $1.5 billion ± $50 million, adjusted gross margin of 43.5% to 44.5%, and adjusted operating margin of 17.5% to 18.5% with adjusted OpEx of approximately $375 million to $390 million. Relative to Q1, Q2 implies continued growth in revenue at the midpoint but a modest step-down in gross margin versus the 44.7% achieved in Q1. This pattern is consistent with management’s view that Q1 benefited from favorable mix and near-term tailwinds, while Q2 and subsequent quarters remain exposed to supply-driven mix and component-cost variability. A concise set of the most decision-relevant management statements from the call follows, capturing the guidance posture and the underlying drivers: “Demand is incredibly strong, with exceptional order activity in the quarter.” “Backlog has increased by approximately $2 billion this quarter to exit Q1 at approximately $7 billion.” “Nearly all new orders we are taking now will be for fulfillment in fiscal 2027.” “Our revenue in the first quarter would have been higher but for these constraints.” “We expect demand will continue to outstrip supply at least for the next several quarters.” “The price increases… haven’t started to fully kick in until the second half of the year… you should expect those price increases to show up in Q3 and Q4.” “Pure underlying demand, across the board, not driven by… pricing thresholds or anything.” STRATEGY AND PRODUCT ROADMAP SIGNALS The call sharpened the narrative around multiple “waves” of opportunity, with concrete product and timing disclosures that matter for 2027+ optionality more than for FY26 guidance. “Scale across” distributed AI training clusters were positioned as early-stage but rapidly ramping, with 3 hyperscalers cited as deploying Ciena optical solutions for training across distance, and additional clusters/orders noted. RLS was described as the “de facto industry line system standard for cloud providers,” with a 2nd consecutive record quarter for RLS shipments and revenue. The most specific forward product catalyst discussed was RLS Hyper-rail, described as delivering “an order of magnitude increase in fiber density within existing rack footprints,” intended to reduce or avoid costs and complexity tied to adding amplifier huts. Management expects to demo a first prototype at OFC in the coming weeks, expects standardization to begin at the end of 2026, and expects a ramp in 2027. This is structurally important because it suggests (a) continued roadmap differentiation at the line system layer, (b) potential for stronger ASP and margin through higher-density solutions, and (c) sustained hyperscaler design-in durability if Hyper-rail becomes a standard. Inside the data center, management emphasized an “opticalization” trajectory as electrical interconnect approaches physical limits, expecting increasing use of optical co-packaged interconnects and a longer-term shift where coherent optics can overtake IMDD for shorter reaches as bandwidth demands rise. The Nubis acquisition was operationalized through the first product introduction: Vesta 200 6.4T optical engine, described as “the industry’s first high-density, low-power, open ecosystem, pluggable CPO solution,” with samples expected in calendar Q2 2026. Nitro linear redriver technology was also highlighted, expected to reduce power by up to 80% versus AEC-type solutions, with samples also expected in calendar Q2 2026. These are strategically meaningful as they expand TAM beyond WAN/DCI, but they also introduce new competitive arenas with different incumbents, faster innovation cycles, and potentially different margin structures than traditional optical transport systems. Decon (data center out-of-band management) was positioned as a defendable niche where Ciena has a lead due to vertical integration, software integration, and installation capability, with Meta as the initial design partner and technical discussions ongoing with 2 other major hyperscalers. Near-term revenue contribution remains uncertain, but the discussion signals continued hyperscaler engagement beyond core optical transport and deeper embedding in data center operations. COMPETITIVE AND CUSTOMER DYNAMICS The competitive tone in the core WAN business was described as more benign, supported by industry consolidation, while emphasizing that differentiation is driven by next-generation line systems and modem technologies. Management asserted continued market share gains in 2025 and an expectation for even stronger share gains in 2026, attributing this to technology leadership, collaboration, and execution. Inside the data center, management acknowledged a “different set of competitors… some of which are very large,” but positioned Ciena’s advantage as high-speed optical leadership and systems knowledge translated into components and integrated solutions. Customer concentration remains notable. Q1 had 3 >10% customers, including 2 global cloud providers and 1 tier-1 North American service provider with strong MOFN activity. This concentration is consistent with the hyperscaler-led AI cycle, but it implies that pacing, architecture decisions, and procurement behavior at a small number of buyers can drive material variance in quarterly outcomes. The call also introduced “neoscalers” as an expanding set of customers showing momentum, largely MOFN-oriented currently, with management expressing caution on direct build approaches for certain neoscalers due to their financial structures. NON-CORE ITEMS AND NOISE FACTORS Tariffs were addressed explicitly. The Supreme Court ruling on 02/20 striking down IEPA tariffs implemented in March 2025 was stated to be immaterial to results, with a new global replacement tariff regime pending final rates. Management expects an immaterial impact based on current information but indicated ongoing monitoring. This topic appeared intended to reduce uncertainty rather than to signal a material swing factor. Capital allocation remained balanced but growth-oriented. The company exited Q1 with $1.4 billion in cash and repurchased approximately 400,000 shares for $81 million. This level of repurchase is modest relative to the size of the demand/supply investment cycle being described and appears secondary to R&D and capacity expansion priorities. INVESTMENT IMPLICATIONS The call materially strengthened the bull-case structure around AI-driven optical connectivity demand and multi-year visibility, but also reinforced that near-term revenue and margin are supply-governed and therefore execution-risked. The incremental datapoints that matter most for positioning are the backlog step-up to approximately $7 billion, the statement that incremental orders are largely for FY27 fulfillment, and the raised FY26 revenue and margin outlook. These signals indicate that the demand environment is not only intact but accelerating, and that Ciena is positioned as a key supplier in both WAN/DCI and emerging “scale across” training cluster connectivity. In a supply-constrained environment, strong backlog and long lead times typically expand strategic supplier power and create optionality for better “value exchange” (pricing, terms, non-cancellable commitments, and cash conversion improvements), supporting sustained operating margin and EPS power. At the same time, the call flagged several factors that can cap near-term upside or raise the risk profile. Management’s guidance explicitly balances demand with supply, suggesting FY26 remains constrained by component availability and manufacturing ramp. Elevated CapEx and long-term supply purchase commitments increase the risk of unfavorable operating leverage if demand were to decelerate or if customers were to rebalance buffered orders once supply normalizes. The decision to guide to roughly equivalent 1H vs 2H gross margins, despite pricing actions expected to land in Q3 and Q4 and despite an 800 ZR ramp narrative, introduces a conservative posture that could reflect real cost/mix uncertainty. Software weakness (Blue Planet down 22% y/y) and services below estimates suggest the cycle is currently hardware-heavy and may not yet be translating into broader software/automation monetization. Key upside and downside vectors implied by the call are as follows. Upside vectors: Supply unlock upside: Management stated Q1 revenue “would have been higher but for these constraints,” indicating that incremental component availability could translate directly into higher revenue conversion within the existing demand pool. Pricing/value-exchange tailwinds: Price increases on new orders are expected to show up in Q3 and Q4, and management described additional levers beyond price (cash conversion, terms, non-cancellable commitments) that can expand economic capture even if headline pricing remains balanced versus share goals. Scale-across and RLS momentum: Continued ramp in training cluster connectivity with 3 hyperscalers deploying Ciena and RLS delivering consecutive record quarters suggests a potentially durable hyperscaler line-system standard position, with Hyper-rail as a 2027 catalyst. Operating leverage durability: OpEx held roughly flat vs 2025 while revenue scales supports structurally higher operating margin, assuming gross margin remains within guidance. Downside and risk vectors: Backlog normalization risk: Early ordering behavior was a central prior-cycle issue, and while management emphasized improved visibility and terms, the backlog magnitude and FY27 fulfillment timing elevate the importance of cancellation/deferral protections and real-world deployment pacing. Component cost inflation and vendor repricing: Management acknowledged supplier cost pressures and the risk of backlog repricing by component vendors, which can compress gross margin if customer value exchange lags cost escalation. Mix and delivery variability under constraints: Supply-driven allocation can skew deliveries toward lower-margin configurations and introduce quarter-to-quarter gross margin volatility, consistent with Q2 gross margin guidance stepping down from Q1’s 44.7%. Competition and execution risk in “inside the data center” markets: The Nubis-driven roadmap (Vesta, Nitro) expands TAM but competes in faster, silicon-centric ecosystems where incumbents are large and product cycles are compressed, potentially pressuring margins and increasing R&D intensity. Software monetization uncertainty: Blue Planet decline suggests software is not currently scaling with the hardware cycle, which may limit multiple expansion if the market expects a broader platform/software mix shift.
Adding $CIEN leap calls. https://t.co/ijNxeIHQ3D
$CIEN (Bloomberg) -- (Adds shares.) Ciena shares are down 3.5% in premarket trading, after the communications equipment company reported its first-quarter results and gave an outlook. •Shares up nearly 50% this year, as of last close FIRST QUARTER RESULTS •Adjusted EPS $1.35 vs. 64c y/y, estimate $1.17 (Bloomberg Consensus) •Revenue $1.43 billion, +33% y/y, estimate $1.4 billion ◦Networking platforms revenue $1.15 billion, +40% y/y, estimate $1.09 billion ◦Converged Packet Optical revenue $1.02 billion, +41% y/y, estimate $970.2 million ◦Routing and switching revenue $126.0 million, +35% y/y ◦Platform software and services revenue $93.3 million, -1.9% y/y, estimate $94.5 million ◦Blue Planet Automation software and services revenue $20.4 million, -22% y/y, estimate $32.6 million ◦Global Services revenue $164.1 million, +26% y/y, estimate $178 million ◦Maintenance support and training revenue $87.6 million, +17% y/y, estimate $81.7 million ◦Installation and deployment revenue $67.9 million, +42% y/y, estimate $72.7 million ◦Consulting and network design revenue $8.6 million, +12% y/y, estimate $17 million •Adjusted gross margin 44.7% vs. 44.7% y/y, estimate 43.5% SECOND QUARTER FORECAST •Sees adjusted gross margin 43.5% to 44.5%, estimate 42.9% COMMENTARY AND CONTEXT •Raising Rev. Guidance Range •Providing revenue guidance of $1.5 billion plus or minus $50 million for fiscal second quarter 2026
These 6 layers are responsible for the immense AI boom that we have seen… 1. AI Compute & Chips ~ $NVDA, $AMD, $ASML, $ARM, $AVGO These companies design and manufacture the processors that AI models run on. 2. AI Compute Operators ~ $IREN, $NBIS, $CIFR They operate large scale infrastructure that delivers AI compute capacity. 3. AI Applications ~ $PLTR, $SNOW, $NOW They sit at the top of the AI cycle and turn data into real world value. 4. AI Security ~ $MSFT, $CRWD, $PANW They turn AI systems, data, and infrastructure from cyber threats. 5. Cloud Platforms ~ $MSFT, $GOOGL, $AMZN, $ORCL They turn AI infrastructure into scalable compute capacity. 6. AI Networking and Connectivity ~ $ANET, $MRVL, $CIEN They move massive amounts of data between servers, racks, and data centers so AI systems function. Without these layers AI would not have scaled as rapidly as it did to reach today’s capabilities…
$CIEN to Acquire Nubis Communications. Ciena has traditionally focused on data center interconnect (DCI) solutions. It is now aiming to extend its reach into intra-data center networking through the acquisition of Nubis Communications. Nubis should bring some expertise in CPO, NPO, and ACC. Market: With this acquisition, $CIEN aims to move beyond the DCI segment into the fast-growing market for intra-data center optical connectivity—which is vital for AI cluster scaling and rack-to-rack communication. IPs: Ciena should gain access to Nubis’s high-bandwidth, low-power optical interconnect IP, helping accelerate the commercialization of Nubis’s emerging products and strengthen Ciena’s position in optical networking.
