Who mentions $DRAM most on Surfaced?

@tradexwhisperer (Trade Whisperer) mentions $DRAM most — 353 mentions in the past 12 months.

When was $DRAM first mentioned on Surfaced?

The earliest tracked mention of $DRAM in the Surfaced archive was on Apr 2, 2026.

How many voices track $DRAM?

36 FinTwit voices have mentioned $DRAM 734 times in the past 12 months.

What kind of voices mention $DRAM most?

$DRAM is mentioned most by Semis voices on Surfaced.

$DRAM — FinTwit mentions on Surfaced

Recent calls

latest 100 of 169

@institlpgpJul 17, 2026 · 5:13 PM

tellin you guys here and now, this tech selloff is about openai delaying it's ipo. $dram as a proxy peaked june 22; openai delay went public within the next 48 hours. and in that regard it is VERY much about glm, kimi, nemotron, tml, etc.

Open on X

@itsmichaelluuJul 17, 2026 · 4:42 PM

Perfect day to add $DRAM calls for Jan 2028 $100 for $14.

Open on X

@yasutaketinJul 17, 2026 · 7:34 AM

メモリー銘柄の将来の需給に不安を抱いている人へ。 $MU $DRAM ヒューマノイドロボット向け計算基盤であるNVIDIA Jetson Thorでは、「128GBのLPDDR5X、帯域273GB/秒」を搭載している。従来の産業ロボットというより、小型AIサーバーを1台ずつ載せる構造に近い。メモリー需要はAIサーバー以外でも爆発すると思うよ。 DRAMもNANDも。

Open on X

@yasutaketinJul 16, 2026 · 11:53 PM

【CXMT大型IPOは、米韓メモリ株の本当の脅威か？】（ChatGPT） $MU $DRAM $SKHY 結論：懸念の方向性は妥当。ただし「すぐ供給過剰になる」は誇張。・CXMTは大型IPOで最大約98億ドルを調達予定・資金は増産だけでなく、DDR5・LPDDR5X・先端DRAM開発にも使われる・IPO資金がすぐ大量供給に変わるわけではなく、工場建設・歩留まり改善・顧客認証には時間が必要・2026～27年のDRAM需給はなおタイトで、短期的な供給過剰リスクは低い・一方、CXMTは政府支援を背景に、採算よりシェア拡大を優先する可能性がある・中長期では、PC・スマホ向けの汎用DRAM価格を押し下げる構造リスクになる・HBMではSK hynixやMicronとの差が大きく、現時点の直接的な脅威は限定的・Micronはメモリ専業色が強いため、株価が最も反応しやすい・Samsungは汎用DRAMへの影響が大きいが、事業分散で株価影響は薄まる・SK hynixはHBM比率が高く、CXMTへの短期耐性は相対的に高い・SanDiskなどNAND株への直接影響は小さく、売られる場合はセクター全体のリスクオフ要因要するに、「短期業績悪化の材料」というより、「2028年以降の汎用DRAM価格を圧迫する中長期リスク」として見るのが妥当。

Open on X

@tradexwhispererJul 15, 2026 · 4:54 AM

$MU "Our view is consistent with management, that AI will keep DRAM demand well above supply BEYOND 2027" -Morgan Stanley Holy Structural $DRAM Shit

Open on X

@photonbullJul 13, 2026 · 11:12 PM

Fuck the Kospi We ride US stocks $DRAM

Open on X

@tradexwhispererJul 13, 2026 · 6:55 PM

$MU $SKHY $DRAM Just Remember HBM pricing is expected to nearly TRIPLE by Q4 2027. Not stabilize. Not hold. Triple. Volatility is driven by fear. The thesis is driven by data. https://t.co/WvfQMowb3o

Open on X

@moninvestorJul 13, 2026 · 4:26 AM

No idea what's happening in Korea. This is exactly why, when it comes to memory stocks, I'd rather own a concentrated ETF like $DRAM than try to pick individual winners. Even though the ETF is down around 9%, I'd still argue it's the safer approach. Most memory stocks have had incredible runs over the last year and trying to identify which one will continue to outperform from here isn't easy. With a concentrated ETF, you still get meaningful exposure to the companies benefiting from the AI memory boom, but you reduce the risk of being wrong on a single stock.

Open on X

@rubicon59Jul 12, 2026 · 7:32 PM

Key memory secular tailwinds: $DRAM 1. Growth in “users” is no longer related to the number of compute device/people 2. Unlike in the past, the benefit of more memory doesn’t top out but grows practically indefintely 3. AI memory is increasingly less and less a swappable commodity and more and more a co-designed ASIC

Open on X

@moninvestorJul 12, 2026 · 3:14 PM

Buy $DRAM

Open on X

@yasutaketinJul 11, 2026 · 4:23 PM

SKハイニクス $SKHY $DRAM ・韓国株のドル換算価値：約144.8ドル・米国ADR：168.0ドル → 米国ADRが約16%割高日本の証券会社でSKハイニクス買えなかったとしても、 $DRAM 買わない手は無いのでは？と思った。落ちてる小銭では。 https://t.co/9Nx80zABAT

Open on X

@itsmichaelluuJul 11, 2026 · 1:50 PM

My favorite one to get right now is $DRAM for January 2028 $100 calls for memory for these 3 reasons: 1/ Memory is THE bottleneck of AI not compute, memory. HBM is effectively sold out, DRAM contract prices ripped ~90% in Q1 alone, and TrendForce called it the craziest memory market in the industry's history. Shortages this structural don't clear in a quarter. 2/ $DRAM is the only pure-play basket for it. solana:SKHYhSjuRWHgikq8eRKbtBbpABgJSkd7ytQV14i9EQ3 , $MU and Samsung make up ~73% of the fund the exact names winning the HBM cycle so you ride the whole supercycle in one ticker without living or dying on a single earnings print. 3/ The Jan '28 LEAP is defined-risk leverage on that thesis. This ETF went from $28 to $81 in its first two months it moves. 18 months of runway, downside capped at the premium, and real convexity if the cycle carries it back through $100.

Open on X

@yasutaketinJul 10, 2026 · 8:27 PM

"ディラン・パテル氏は、今回のメモリ不足は通常のサイクルではないと述べている・今後3年間、メモリ容量は年20〜30%しか成長しないのに対し、需要は倍増し続けているため、メモリ価格は急騰し続ける。・今回のメモリ不足は短期的なものではなく、何年も続く見通しである。・メモリ価格はすでに4倍に上昇しており、容量の成長に加えてさらに2〜3倍上昇する。・推論によってKVキャッシュが爆発的に増加するため、メモリが最大の勝者になると指摘している。・メモリの売上高総利益率はまだ85〜90%には達していないが、いずれ到達し、その後ある時点で70%台かそれ以下へと半減する。" $DRAM $MU

Open on X

@tradexwhispererJul 10, 2026 · 4:55 PM

$MU $DRAM $SKHY Apple squeezed memory makers on pricing for a decade. That era is OVER. 2023 → 2030 Mobile Share: 40% → 18% Data Center Share: 30% → 62% The biggest power reversal in memory history. HBM eats 3x the wafers. AI sets the price now. Fuck Apple.

Open on X

@kevinvestingytJul 9, 2026 · 11:40 PM

For the record, I expect memory to continue rallying tomorrow. $SNDK $MU $DRAM

Open on X

@tradexwhispererJul 9, 2026 · 4:21 PM

$SKHY $DRAM “Chairman Choi will attend the NASDAQ listing ceremony in New York on the morning of the 10th (local time) along with key executives including President Kwak No-jung of SK Hynix.” “In the business community, there is speculation that during his U.S. visit, Chairman Choi will meet successively with executives from major big tech companies such as NVIDIA and Tesla to discuss ways to expand cooperation in AI memory.” “The scale of SK Hynix’s ADR is approximately $31.5 billion USD (≈43 trillion KRW), with up to 17.79 million new shares (about 2.5% of total issued shares) to be issued.” “The raised funds will be used for the construction of the Yongin Semiconductor Cluster Phase 1 Fab, Cheongju P&T7 Advanced Packaging Fab, and acquisition of equipment including EUV scanners.” “Chairman Choi’s direct participation in this event is interpreted as a move to actively promote SK Hynix’s AI memory competitiveness and growth potential to global investors.”

Open on X

@midascabalJul 8, 2026 · 11:44 PM

$DRAM gives you exposure to the entire memory supercycle. Diversified across the sector. Memory is the constraint. TAM expansion beats company risk.

Open on X

@yasutaketinJul 8, 2026 · 5:46 AM

メモリー需要拡大はLLMだけじゃない👇 $DRAM $MU 【LLMにロボティクス・自動運転が重なり、メモリー需給はさらにタイト化へ】（ChatGPT）・メモリー半導体の需給は、今後もタイト方向に向かう可能性が高い・主因は引き続きLLM／AI推論インフラ　HBM、サーバーDRAM、eSSD、高性能NANDの需要が強い・加えて、ロボティクスはオンデバイスAI化でメモリー搭載量が増える　ロボット本体で画像・言語・行動モデルをリアルタイム処理するため・自動運転／ADASも車載メモリー需要を押し上げる　カメラ、LiDAR、レーダー、地図、車内AI、インフォテインメントでDRAM／NAND使用量が増加・重要なのは、HBMがDRAM供給能力を食う構造　HBMはウェハを多く使うため、汎用DRAMの供給余力を圧迫しやすい・NANDもAI推論、RAG、ベクトルDB、ログ保存、モデル保存で需要が広がる　AIブームはHBMだけでなく、eSSD／高容量NANDにも波及・ただし短期の主役はまだAIデータセンター　ロボティクス・自動運転は2027年以降の中長期ドライバーとして見るのが妥当・恩恵が強そうな順番は、　HBM ＞サーバーDRAM ＞ eSSD／高性能NAND ＞車載メモリー＞ロボティクス向けメモリー結論：メモリー需給のタイト化は、単なる一時的サイクルではなく、AIインフラ＋ロボティクス＋自動運転による構造的な需要増として見た方がよさそう。

Open on X

@tradexwhispererJul 7, 2026 · 6:21 PM

$MU $DRAM LPDDR pricing could see hikes exceeding 20% for Q3 2026. Apple is Cooked. SOCAMM2, the new AI server memory standard built on LPDDR5X, is being adopted across the board. Nvidia's Vera Rubin platform uses it. AMD is building it into future Instinct GPU platforms. Qualcomm is exploring it for its AI200 and AI250 accelerators. LPDDR is no longer just phones and laptops. It is now the memory of choice for AI servers across three major chip platforms, all chasing energy efficiency and all competing for the same constrained supply. HBM eats DDR wafers. SOCAMM eats LPDDR wafers. Bullish 🔥

Open on X

@alphaticaioJul 7, 2026 · 6:04 PM

$MU this is going to move soon. The Block Brigade is making a comeback. They are buying the dip. Do not be fooled by the crazy talk on this platform. $MU $DRAM https://t.co/daDNV4lT28

Open on X

@tradexwhispererJul 6, 2026 · 6:01 PM

$MU $DRAM $SKHY Holy Structural Shit DRAM Prices will climb for the next ~2 years due to Chronic Undersupply. Every quarter a new record. Q4 2027 alone is projected at ~$260 billion in revenue at 86%+ margins, so roughly $1 trillion in one year. https://t.co/7yE24D13Wm

Open on X

@ckcapitalxxJul 6, 2026 · 4:09 PM

Reminder Samsung Earnings are today. Massive day for $DRAM.

Open on X

@tradexwhispererJul 5, 2026 · 4:41 PM

$MU $DRAM Holy TRIPLE Shit. Shorts are Fucked. HBM ASP goes from $1.26 in Q1 26 → $3.25 by Q4 27. That is nearly a TRIPLE. Non-HBM keeps climbing too. NO TOP IN SIGHT. Holy Structural Shit! https://t.co/qHwBLtpTlZ

Open on X

@yasutaketinJul 5, 2026 · 10:32 AM

【キオクシア高速SSDとSNDKのHBF、何が違う？】 $285A $SNDK $DRAM ・結論：どちらもAI推論時代の「HBM不足」を補う技術だが、立ち位置が違う・キオクシアの高速SSD → SSDを超高速化して、GPUの外側に置く「高速倉庫」・SanDisk・SKハイニクスのHBF（High Bandwidth Flash） → NANDをSSDではなく、HBMに近い「大容量メモリ層」として使う構想・キオクシアは「SSD寄り」 → GPUが直接アクセスできる超高IOPS SSD → KVキャッシュ退避、RAG、ベクトルDB、長文コンテキスト向け・HBFは「メモリ寄り」 → HBMとSSDの中間に置く新階層 → 巨大モデルの重みや推論データを、HBMより安く大容量で支える狙い・たとえるなら、 HBM＝机の上の資料 HBF＝机の横の大型ファイル棚キオクシア高速SSD＝部屋の中の高速倉庫通常SSD＝別室の倉庫・投資目線では、キオクシアは「AI向け高付加価値SSD」路線・SNDKは「NANDをAIメモリ階層に昇格させる」路線・つまり、キオクシアだけが先行しているというより、キオクシア＝SSDをAI用に高速化 SNDK＝NANDをHBM補完メモリ化という別アプローチ・OpenAIのような推論効率化が進んでも、モデル保存・重みロード・KVキャッシュ・長文処理の需要は残る・そのため、AI推論効率化はGPU需要には警戒材料だが、SSD/NAND企業への直接ダメージは比較的小さい可能性がある

Open on X

@tradexwhispererJul 4, 2026 · 1:57 AM

$DRAM $EWY $SKHY “South Korea’s exports surpassed $100 billion in a single month for the first time in June, driven by booming semiconductor shipments. It jumped 70.9 percent from a year earlier to a record $102.25 billion.” “Semiconductors led the surge as soaring memory chip prices, fueled by strong demand tied to the global artificial intelligence infrastructure boom, lifted exports 199.5 percent from a year earlier to a record $44.82 billion. It was also the first time monthly semiconductor exports exceeded $40 billion.” Computer exports surged 308.8% to $5.41 billion. Wireless communication device exports climbed 51.9% to $1.55 billion. 18 of Korea’s 20 major export categories posted year-on-year increases. Exports to China and the U.S. both exceeded $20 billion. Trade surplus reached a record $36.15 billion in June (first time over $30 billion); first-half surplus $138.3 billion. First-half exports: record $496.7 billion, on track for annual exports to exceed $1 trillion for the first time.

Open on X

@tradexwhispererJul 3, 2026 · 5:18 PM

$MU $DRAM Memory & Storage are dead. Those words belong to the era of spreadsheets & PowerPoint. What sits next to a GPU in an AI Supercomputer is not "memory." It's the substrate of thought. This is the Cognitive Capacity, the Thinking Layer. Time to call it what it is $SKHY https://t.co/4cSNMDg6SW

Open on X

@tradexwhispererJul 2, 2026 · 6:48 PM

$MU $DRAM Priced for a downturn. Set to deliver record profits in 2027. Supply-demand to remain tight beyond calendar 2027.

Open on X

@rubicon59Jul 1, 2026 · 9:46 PM

I am posting this (in plain English) so I can quote it later. Memory, the “gold” of AI, is selling off at single digit multiples. It takes extra high IQ to mess it up. $DRAM

Open on X

@institlpgpJul 1, 2026 · 2:45 PM

meta has a 6gw pipeline of amd gpu coming if he can make mi450 work, he's arbing the gpu market this is massively bullish compute meta sees the gpu rents xai got from google and anthropic and is selling into it zuck is economically rational adding $dram and $meta here

Open on X

@bullofbritainJul 1, 2026 · 9:14 AM

@Timing_Markets If you cant choose. $DRAM is your friend

Open on X

@daniel_kossJul 1, 2026 · 8:29 AM

This is exactly what I told Subscribers Yesterday will be a main catalyst for $CCXI Roundhill (famous for their $DRAM ETF) added $CCXI to their Humanoid Robotics ETF. I wrote to Subscribers: "Re $CCXI, I think we might see insane FOMO into this name soon. Don't think real move is over. I simply expect huge amounts of money to chase this theme over the next 12 months. Huge capital inflows, and there’s very little they can buy that’s this legit, with customers, investors, and partners like NVIDIA, Amazon, Mercado Libre, GXO, Toyota, Schaeffler, Foxconn, SoftBank, DCVC, and Playground Global." This is VERY similar to $OKLO which had a 1800% run! That's why I said $CCXI is the ideal way to retardmaxx Humanoids. US based and US listed humanoid pure play. Billions getting forced into this space via ETFs over the next 2 years. There's just so much buying pressure. Oklo went up 1800% largely because it was part of every Nuclear ETF, I believe. Not financial advice. Just retardmaxx thoughts.

Open on X

@itsmichaelluuJul 1, 2026 · 12:57 AM

This is possible in 1 year's time if $DRAM hits $400 and $MU between $3000-$5000. Yes, you'd 10x your trade, but if not you'd still make a crap ton of money. What do you think? Step-by-step: 1. One call costs about $2,500. 2. To make $25,000, that option needs to trade near $250 per share. 3. Since the strike is $100, $DRAM likely needs to trade around: $100 strike + $250 option value = ~$350 DRAM Current DRAM is around $75. So the real target is: $75 → $350 That is about a 4.6x move in DRAM. For that to happen, $MU probably needs a monster memory-cycle move too. Rough estimate: MU likely needs to be around $3,000–$5,000+ Why? Because $DRAM is not just Micron. It is a basket of memory/AI storage stocks. MU going parabolic helps, but the whole memory sector likely needs to rerate. The trade: Buy 1 Jan 2028 $100 DRAM call for ~$2,500. Bull case: $DRAM to $350 Option value near $250 Contract value near $25,000 Risk: If DRAM stays below $100 by expiration, the call can expire worthless. That’s the game: Defined risk: $2,500 Target reward: $25,000 Needed move: DRAM to ~$350 Likely MU level: $3,000–$5,000+ Big risk. Big reward. Memory supercycle or zero.

Open on X

@tradexwhispererJun 30, 2026 · 5:09 PM

$MU $DRAM $SNDK 6th type of idiot showed up today: "China will flood the memory supply." Flood it with what? A drought? Sanctions have cut off access to cutting edge equipment. They can't make HBM. Can't make SOCAMM. No Advanced Packaging. No EUV access. No path to leading edge nodes. No advanced AI server memory at scale. Meanwhile, Alibaba, Tencent, ByteDance, and Huawei Cloud are all actively signing long term agreements with Samsung, SK Hynix, and Micron for HBM and advanced server DRAM. Domestic supply cannot support their AI training clusters. If China could flood the market, their own cloud providers wouldn't be raining in Korea and the US. I've spoken

Open on X

@jukan05Jun 30, 2026 · 8:35 AM

TrendForce: 3Q26 PC DRAM contract price growth has been revised up from the previous forecast of +8–13% QoQ to +15–20% QoQ. The 4Q26 forecast was also revised up from +0–5% to +3–8%. 3Q26 server DRAM contract prices are expected to rise by 13–18% QoQ. $DRAM $MU

Open on X

@photonbullJun 29, 2026 · 7:36 PM

Korea about to take $DRAM $SNDK $MU 5% higher overnight btw

Open on X

@bullofbritainJun 29, 2026 · 3:37 PM

Im losing less and less money everyday. This can only mean bottom is in soon. NFAa $DRAM

Open on X

@rubicon59Jun 29, 2026 · 2:50 AM

Building #Terafab suggests Musk does not expect sufficient AI compute and memory capacity to be readily available in the medium to long term. That stands in sharp contrast to the view that AI memory supply will soon catch up with demand. $DRAM

Open on X

@tradexwhispererJun 29, 2026 · 2:17 AM

$MU $DRAM $SNDK Holy Shit Now even Chinese CSPs are lining up for Long Term Agreements. A small excerpt from what I shared on Patreon. From a private industry research firm: "Most US CSPs have completed their negotiations on LTAs for either 2026 to 2028 or 2026 to 2030 in 2026, while Chinese CSPs are looking to conclude their LTA talks between late Q2 and Q3 2026, thereby further carving up available capacity among suppliers." Both sides of the Pacific. LTAs. Boom!

Open on X

@photonbullJun 28, 2026 · 10:46 PM

The next memory trade is still memory As long as the stocks are still trading at low multiples and the demand remains, there is no need to overcomplicate it Long $MU $SNDK $DRAM

Open on X

@rubicon59Jun 28, 2026 · 10:41 PM

Forget about memory supply catching up to demand. The gap is in fact widenng. $DRAM

Open on X

@tradexwhispererJun 28, 2026 · 4:21 PM

$MU $DRAM $SNDK Nobody is talking about Micron's Automotive/Embedded segment. 71% QoQ growth. $2,708M to $4,634M in one quarter. L2+ vehicles require 5x the memory of an average vehicle. That mix is more than doubling this year to 20%+ and expected to exceed 40% by 2030. Now layer this on top: humanoid robots carry 10x the memory of an L2+ vehicle. That demand cycle begins later this decade and SUSTAINS for multiple decades. 5x for autonomy. 10x for robotics. Physical AI Incoming. Buckle Up

Open on X

@kevinvestingytJun 28, 2026 · 12:19 AM

This "memory crisis" likely gets so much worse than 99.9% of consumers can imagine over the next few years. I've been telling all my close friends/family to buy their new computers/laptops now if they've been thinking about getting one. The only hedge is going long $DRAM $MU etc

Open on X

@rubicon59Jun 27, 2026 · 9:57 PM

First they laughed at memory makers. Then they vlified them them. Later started signing 3-yr contracts, now 5-yr. How long before 8-10yr contracts are the norm secure capacity? $DRAM

Open on X

@tradexwhispererJun 27, 2026 · 3:41 PM

$MU $DRAM $SNDK Bank of America just put out their 2030 semi TAM. $1.96T total. $900B is memory. Nearly half. And this is almost certainly outdated. It's not pricing in the Agentic AI boom yet. https://t.co/XEWklLCg4L

Open on X

@itsmichaelluuJun 27, 2026 · 2:37 PM

If $MU crashes below $1000 or $900 or $800. I like to do these 3 things: 1. Add shares and hold until $2000+ 2. Buy $DRAM calls for January 2028 $100 calls 3. Buy $MU calls for January 2028 $2000 calls The key now is to buy as much dips as possible and wait. https://t.co/JQ8wm6Poo2

Open on X

@photonbullJun 27, 2026 · 8:17 AM

Could this be a DeepSeek moment for memory stocks👀 We know what to do here if it dips more $MU $DRAM $SNDK https://t.co/yVJ4Eev0ju

Open on X

@bullofbritainJun 27, 2026 · 7:50 AM

If $DRAM are going to start including Chinese equities as the norm, its probably better r/r than holding any single memory name

Open on X

@tradexwhispererJun 26, 2026 · 6:00 PM

$MU Holy Shit. The structural HBM ripple effect. HBM direct revenue by 2031: $1.4T Non-HBM incremental revenue: $3.6T Total $DRAM impact: $5.0T The non-HBM revenue is 2.5x larger than HBM itself. HBM isn't just a product. It's a pricing engine for all of DRAM. https://t.co/jnEGDWQePk

Open on X

@rubicon59Jun 26, 2026 · 2:24 AM

Better and richer context alone can dramatically increase model intligence without any advancement of the model itself. And more memory is pretty much synonymous with better context. $DRAM demand.

Open on X

@firstadopterJun 25, 2026 · 4:54 PM

WAT! Microsoft: "Unfortunately, console storage and memory prices have increased by more than 2.5x and we expect another doubling by the fall of 2027." $DRAM $MU

Open on X

@alphaticaioJun 25, 2026 · 1:00 PM

Wall Street still has room to hike price. Barclays is the only firm that went above $1,900 this morning. They raised their $MU price target from $1,175 to $2,000. $DRAM $MU

Open on X

@rubicon59Jun 25, 2026 · 6:42 AM

We are not far from the memory-business narrative flipping from mah cyclical to mah growth. IMHO $DRAM

Open on X

@moninvestorJun 25, 2026 · 6:38 AM

Buy $DRAM.

Open on X

@tradexwhispererJun 25, 2026 · 5:57 AM

$DRAM Holy Shit SK Hynix up 15%+ Nasdaq 100 Futures up 2% $MU is the true AI leader. Not a commodity. Not a debate anymore. https://t.co/2vFRZwi2om

Open on X

@alphaticaioJun 24, 2026 · 9:04 PM

@The_WholeArmour We were hedged with $DRAM

Open on X

@tradexwhispererJun 24, 2026 · 4:11 PM

$MU $DRAM $SNDK The most powerful CEOs on earth are telling you memory is in a structural transformation. 12 most important quotes. "The future of artificial intelligence will be shaped as much by memory as by computing power." -Jensen Huang "Go ahead and build memory fabs; however much capacity you add, NVIDIA will consume it." -Jensen Huang "The whole industry supply chain, everything from wafers to packaging to silicon photonics... everything's in short supply because the demand is so high. It is going to persist for several years." -Jensen Huang "The entire supply chain is challenging this year because demand is so much more. Supply has been growing 100% every year, but demand is going faster than that." -Jensen Huang "I think it will be for quite a few years to come... Demand is so massive." -Jensen Huang "I would say memory in advanced silicon, it's kind of the 25 times 25... From 80 gig to 2 terabytes is 25 times more memory per accelerator. And in that timeframe, you'll have about 25 times more accelerators. So 25 times 25 is 625 times." -Michael Dell (Dell CEO) "Demand is in excess of the supply by quite a lot... So we have a severe shortage in memory silicon and also in some other areas." -Michael Dell (Dell CEO) "As memory per accelerator and system scale expand simultaneously in AI infrastructure, a structure is forming where total memory demand increases by approximately 625 times. We are still in the early stages of technology adoption." -Michael Dell (Dell CEO) "[US Expansion] is a tiny fraction of the memory that's needed, and in fact, even if you take the best case assumptions of the memory makers... it is not enough to meet the demand that is anticipated." -Elon Musk "We see that tightness continuing into 2027, so we see durable industry fundamentals over the foreseeable future, driven by AI demand." -Sanjay Mehrotra (Micron CEO) "AI actually wants to have a lot of HBM, and once you make the HBM...we have to use a lot of wafers... The current shortage could continue until 2030, so we expect more than a 20% shortage of the wafers." -Chey Tae-won (SK Group Chairman) "Without the HBM memory, there is no AI supercomputer." -Jensen Huang What exactly are you waiting for?

Open on X

@photonbullJun 24, 2026 · 1:01 AM

$DRAM $SNDK $MU 10% days incoming across the memory names https://t.co/Eh7X3gOCVe

Open on X

@rubicon59Jun 23, 2026 · 10:28 PM

In the long run, we are all doomed (maybe). Might as well enjoy the prosperous now and the short term (a sure thing). $DRAM

Open on X

@kevinvestingytJun 23, 2026 · 7:20 PM

Which stocks did you guys buy/sell today? I bought a lot of $MU $DRAM

Open on X

@tradexwhispererJun 23, 2026 · 3:48 AM

$DRAM $MU On Jun 22nd, Park Jun-young, an analyst at Hanwha Investment & Securities, raised the PT for SK Hynix to 4.3 million won. This represents 55.6% upside. Researcher Park explained, "Korea's memory industry is overcoming its past weaknesses based on powerful weapons such as LTA and HBM" In other words, SK Hynix is no longer a highly cyclical memory company; it can now sustain high profits consistently. No longer justifies low valuation and should be valued more like global tech peers. Exactly what I said in Jun 13th.

Open on X

@alphaticaioJun 23, 2026 · 12:27 AM

Predicable $MU pre-earnings behavior. This trade is not done yet. Don't panic. $DRAM https://t.co/bDurDAUb7V

Open on X

@tradexwhispererJun 22, 2026 · 6:14 PM

$MU $DRAM Holy Shit Micron and Anthropic Announce Strategic Agreement to Scale Next-Generation AI Infrastructure Multi-year memory and storage supply agreement covering Micron’s full data center portfolio, including high-bandwidth memory (HBM), DRAM, and SSDs. Collaboration on AI architecture design to optimize performance, efficiency, cost, and scaling for Anthropic’s Claude models (training and inference). Micron is making a strategic investment in Anthropic’s Series H funding round. Additional elements include enterprise adoption of Claude at Micron and joint analysis of memory/storage performance across AI workloads. The partnership underscores that memory chips are now critical, specialized AI hardware. The Cognitive Capacity Layer.

Open on X

@photonbullJun 22, 2026 · 3:08 PM

Everything is being sold off to buy more memory stocks And with the incredible growth, margins and projections, I doubt anyone can blame them $MU $SNDK $DRAM https://t.co/6xejgzm1Jt

Open on X

@kevinvestingytJun 22, 2026 · 1:55 PM

You’re not too late for the memory trade $DRAM $MU

Open on X

@rubicon59Jun 22, 2026 · 1:51 PM

Includes a multi-year suppky agreement. Another signal that storage is becoming a strategic AI component rather than a commodity. $DRAM

Open on X

@firstadopterJun 22, 2026 · 1:10 PM

MULTI-YEAR GROWTH SUPPLY AGREEMENT $MU $DRAM "Building on the technical collaboration, Micron and Anthropic have entered into a memory and storage supply agreement spanning Micron's industry-leading data center portfolio. This positions Micron to support Anthropic’s multi-year growth trajectory as the frontier AI lab scales its compute strategy for the long term."

Open on X

@tradexwhispererJun 21, 2026 · 6:54 PM

$MU Stifel TRIPLED their old PT ahead of earnings. Three of them set $1,500 targets. Re-rate Incoming. We are just getting started. $DRAM https://t.co/d0T1Vppxva

Open on X

@moninvestorJun 20, 2026 · 7:43 AM

@Ashton_1nvests Why? $DRAM is a great ETF.

Open on X

@tradexwhispererJun 20, 2026 · 2:45 AM

$MU +66%. $SNDK +63%. In one month Last month you asked "what is next." Next month you will ask again. My answer has not changed since my Bargain of the Century call at $62. $DRAM 9x forward. Everything so far was just a warm up. https://t.co/MISafAsFAN

Open on X

@photonbullJun 19, 2026 · 11:22 PM

Long. Memory. It is still the easiest trade/investment here. It did not run up too much. The cycle will not suddenly end with the amount of AI related memory demand. At $300, people said they were looking for the next $SNDK. At $600, they said the same thing. Then at $1000, $1500 and now above $2000. The next $SNDK is $SNDK and the next $MU is still $MU This is not over yet and will not be over for a good while. Plain and simple. $SNDK $MU $DRAM SK Hynix Samsung Kioxia

Open on X

@tradexwhispererJun 19, 2026 · 7:54 PM

$MU $SNDK I respect Charlie Bilello, but comparing $MU today to the dot com era is lazy analysis. Here's the breakdown why Industry Structure 2000: 20+ $DRAM players in a commodity knife fight 2026: 3 player oligopoly with pricing power and long term agreements Product Mix 2000: Commodity DRAM for PCs 2026: HBM, SOCAMM, NAND, DRAM across a diversified portfolio End Market 2000: PC dependent 2026: AI Data Center, Automotive, Graphics, Physical AI Business Model 2000: Spot pricing, zero visibility, 2026: Up to 5-year Long term contracts with hyperscalers, structural demand floor

Open on X

@tradexwhispererJun 18, 2026 · 5:20 PM

$MU $DRAM Holy Structural Shit "HBM sales will add $1.4T in DRAM revenue by 2031 and an incremental $3.6T in non HBM revenue due to higher pricing from HBM related supply constraints." The ripple effect matters more than HBM itself. -June Memory Research Report.

Open on X

@rubicon59Jun 18, 2026 · 4:21 PM

Samsung Electronics reports 57.3 trillion WON ($38b) in operating profit for the first quarter, surpassing Microsoft, Amazon, and Meta. $DRAM

Open on X

@ckcapitalxxJun 18, 2026 · 3:37 PM

$DRAM continues to hit new all time highs and the stocks inside it are still cheap. That sounds like a contradiction. It is not. The memory makers have run hard, but their earnings have run harder, which means on a forward basis they are still trading like bargains. Look at the actual multiples. SK Hynix went up more than 9x in a little over a year and still trades around 5 to 7 times forward earnings. Micron crossed a $1 trillion market cap with a trailing P/E in the 40s but a forward P/E around 9, because analysts know the earnings explosion is just getting started. Samsung screens as outright undervalued. These are not stretched valuations. They are cyclical-looking multiples on companies entering the least cyclical stretch of demand they have ever seen. And here is what changed that the market has not fully repriced. Memory makers just moved from one-year contracts to three and five-year strategic agreements. That is enormous. The entire reason memory stocks always traded at low multiples was the boom and bust cycle. Lock in multi year deals with hyperscalers and that volatility flattens out. Stable, predictable, contracted revenue earns a premium multiple. The market has not paid for that re rate yet. Stack it up. HBM sold out through 2026 and booked into 2027 and 2028. The shortage not expected to ease until 2028. Demand under multi-year contracts now. And forward multiples still sitting at single digits to low teens. The all time high in $DRAM is not the top of the move. It is the market slowly waking up to a re-rate that has barely started. You own SK Hynix, Samsung, and Micron together in one ticker, right as the cheapest part of the AI trade gets repriced. There is a lot of runway left here.

Open on X

@itsmichaelluuJun 18, 2026 · 2:17 PM

@meme01427 Buy $DRAM Jan 2028 $105 calls its at $24 right now.

Open on X

@tradexwhispererJun 17, 2026 · 10:28 PM

$MU $DRAM $TSM just flagged a major shift in AI chip design priorities. Energy efficiency is now more critical than raw compute power. Surging electricity demand from AI applications is forcing the entire industry to rethink the equation. This is not a future trend. It is already the buying criteria. By next year every major procurement decision will have one non negotiable requirement. The new chip must be more energy efficient than what it replaces. No exceptions. SOCAMM will be huge next 5 years. Micron was first to this with SOCAMM. Built for energy efficient AI inference at scale. I've spoken.

Open on X

@ckcapitalxxJun 17, 2026 · 12:47 PM

There is a shortage so severe Goldman called it the worst in 15 years, and most retail investors have no idea it is happening. It is memory. Every company on earth is fighting over NVIDIA chips, but those chips do nothing without memory feeding them, and an AI server eats 8 to 10 times the memory of a normal one. Demand went vertical and the makers cannot keep up. Micron already sold out its entire 2026 production, the whole year gone. Jensen Huang walked up to an SK Hynix wafer at COMPUTEX and wrote “Please Make More” on it by hand. This is not the old memory cycle that booms and busts every couple years. It is structural this time. New fabs take years and tens of billions to build and nothing real comes online until 2027. So demand keeps outrunning supply, and that means years of pricing power for everyone who makes the stuff. HBM for training, DRAM for inference, NAND for storage. Every layer short at the same time. The whole sector is moving on it. SK Hynix, Samsung, Micron, SanDisk all riding the same wave. The catch is the two biggest winners trade overseas and are a pain for most people to buy. That is exactly why I own the $DRAM ETF. One ticker gets me the entire memory complex, every company that wins from the shortage, in a single position. Long Memory.

Open on X

@tradexwhispererJun 16, 2026 · 6:15 PM

$MU $DRAM Where analysts & market got it wrong. They model HBM like it has been around for decades. HBM is a new architecture built to feed AI models context at speed. It did not exist 3 years ago in any meaningful revenue. It is not where data sits. It is where AI thinks.

Open on X

@tradexwhispererJun 15, 2026 · 6:04 PM

🚨 $MU $DRAM $SNDK Important Reminder Huge inflection point for the memory industry. Micron signed its first ever 5 year Strategic Customer Agreement. Not one year. Five. Samsung and SK Hynix are following suit, pursuing 3 to 5 year contracts with major tech firms. For decades, memory has been defined by cycles. That is ending. Multi year commitments bring predictable revenue, operational stability, and structural dampening of the boom bust cycles that have kept PE multiples compressed for 30 years. One year deals gave Wall Street an excuse to call memory cyclical. Five year deals remove that excuse. As revenue visibility extends, PEG ratios move toward 1.0 and PE expansion follows. The market has not priced this in yet. The real breakout has not even happened.

Open on X

@moninvestorJun 14, 2026 · 8:38 PM

If you're scared to buy $SNDK or $MU, just buy $DRAM. It's one of the best ETFs for gaining exposure to memory stocks. The fund is already up 134.19% in just over two months. Absolutely incredible. https://t.co/ADSSey3Wnd

Open on X

@tradexwhispererJun 14, 2026 · 5:29 PM

$MU $DRAM Unbelievable $707,000,000,000. That is the projected 2027 operating profit of 3 memory companies. Apple + Microsoft + Google + Amazon + Meta + Tesla combined? ~$661 Billion. The Memory Trio beats the Magnificent 6. Not revenue. Operating profit. https://t.co/O52OmAUtIA

Open on X

@tradexwhispererJun 13, 2026 · 5:38 PM

$MU $DRAM Wanna know when supply catches up? → HBM eats 30% of all DRAM wafers next year. Possibly 40% by 2028. → That leaves 60% for everything else, Mobile, PC, Server, Automotive. → New fabs are only a fraction of future demand. Never https://t.co/RVihBIC5cZ

Open on X

@photonbullJun 11, 2026 · 11:12 PM

Here is the next big memory catalyst Koreans like to gamble They are obviously going to beat Czechia with a 3-0 score They will have a lot of bets on this They win. They cash out. They buy more memory stocks. Bullish for Samsung SK Hynix $MU $SNDK $DRAM https://t.co/i4inH7ZOiz

Open on X

@tradexwhispererJun 11, 2026 · 5:47 PM

$MU $DRAM HOLY SHIT 85+% margins sustained into 2027. This changes everything. https://t.co/DwssEYOHmx

Open on X

@ckcapitalxxJun 11, 2026 · 5:29 AM

Very bullish for $DRAM

Open on X

@tradexwhispererJun 10, 2026 · 2:36 AM

$MU $DRAM You're right. It's cyclical. Up. Then more up. Then even more up. 250 exabytes to 1,226 by 2031. Nearly 5x. Datacenter growing 34% annually Cute word. Wrong thesis. https://t.co/C8IwP50URf

Open on X

@tradexwhispererJun 9, 2026 · 2:46 PM

$MU $DRAM Bernstein says HBM4 hits $53/GB in 2027. HBM3E is $15/GB today. That is 3.5x in one generation. Now do the math on C-HBM4E, HBM5, and HBM5E for 2028 to 2030. $85/GB? $100/GB? And you still think HBM is a commodity? Intelligence

Open on X

@itsmichaelluuJun 9, 2026 · 11:13 AM

The best one to get would be $MU calls for June 2028 $1500 calls, but this is very expensive. Shorter term ones would be Jan 2027 $1200-$1500 calls. If you can't afford this getting $DRAM options and shares would be the BEST alternative.

Open on X

@tradexwhispererJun 8, 2026 · 7:28 PM

$MU $DRAM Holy Shit. HBM4 prices for 2027 expected to be MULTIPLES higher according to Trendforce. Not percent higher. MULTIPLES higher. 2027 negotiations start NOW. https://t.co/6SKG8VgCVB

Open on X

@firstadopterJun 8, 2026 · 1:27 PM

Bernstein: "Total semiconductor sales grew 106.4% YoY in April after rising 88.1% in March. YoY strength was again dominated by memory with sales almost quintupling at 364.1% YoY" $DRAM https://t.co/lDqRQgxcCz

Open on X

@yasutaketinJun 8, 2026 · 12:48 AM

やっぱKOSPIのサーキットブレイカー発動後がメモリ株の買い場だよなぁ。高値で買うべきではない。一目均衡表の基準線割った後が望ましい。 $MU $DRAM https://t.co/kJL6vRO2Mu

Open on X

@yasutaketinJun 7, 2026 · 10:20 PM

日経、韓国コスピに注目韓国株が止まるなら $DRAM 押し目買いが頭に浮かぶ

Open on X

@ckcapitalxxJun 7, 2026 · 8:42 PM

NVIDIA CEO Jensen Huang just said the memory shortage is going to last for years. He said it directly while meeting with SK Group’s chairman in Seoul. His words. Everything across the supply chain, from wafers to silicon photonics to cable connectors, is in short supply, and demand is enormous. When the most important man in AI tells you a shortage lasts for years, you listen. And I still think the cleanest way to play it is the $DRAM ETF. And the thesis is super simple. This is not a normal memory cycle. For decades memory was a boom and bust commodity. This time it is structural. AI servers need 8 to 10 times the DRAM of a traditional server. Memory makers are pulling capacity away from regular DRAM to build high bandwidth memory for AI accelerators, which only tightens the supply of everything else. Industry fulfillment rates for server DRAM are reportedly below 50%. And new fab capacity does not come online until 2027 at the earliest. A new fab takes 18 to 24 months and tens of billions of dollars to build. The shortage is locked in. Now here is the part most retail investors miss. The two biggest winners of this entire shortage are SK Hynix and Samsung. SK Hynix alone supplies an estimated 50 to 70% of NVIDIA’s HBM4. They are the kings of this cycle. But both trade on the Korean exchange, which makes them difficult for most US retail investors to actually buy. That is exactly why the $DRAM ETF is so useful. It is one of the only clean ways to get real exposure to SK Hynix and Samsung alongside Micron and the rest of the memory supply chain in a single US listed ticker. You own the entire structural shortage in one position. The HBM leaders, the DRAM makers, the equipment names. All of it. Jensen says years. The supply math says years. New capacity does not arrive until 2027. The $DRAM ETF is how I want to own the whole thing.

Open on X

@investmentguru_Jun 7, 2026 · 2:30 PM

HBM demand isn’t slowing down. Memory is the bottleneck. $DRAM gives you a basket of every major player in the space.

Open on X

@jasonschipsJun 7, 2026 · 12:22 PM

One recurring theme that I noticed is how literally every single tray that you see at Computex has a shit ton of $DRAM. It sounds obvious, but you really need to see it for yourself to intuitively grasp how much DRAM is in everything. For example, this is the Vera CPU compute tray. Those SOCAMM modules are stacked too, so it’s actually even more surface area than it appears to be. Being bullish on CPU proliferation for agents is almost just as much bullish on DRAM. As it would drastically increase the total base of compute (accelerated or not) to which memory is attached. I covered this in more detail in my Computex review.

Open on X

@jukan05Jun 7, 2026 · 5:01 AM

TrendForce has revised up its 2Q26 server DRAM contract price forecast to +50–55% QoQ, from its previous forecast of +45–50%. $DRAM $MU

Open on X

@itsmichaelluuJun 6, 2026 · 12:57 PM

This is $MU. It's the greatest bottleneck of AI infrastructure and all its memory is sold out until end of 2027. If you buy 1 stock for the rest of your life, this is the one. And $DRAM too. https://t.co/gc33n9HNn3

Open on X

@tradexwhispererJun 5, 2026 · 5:45 AM

$MU $DRAM The DDR5 shortage is bringing DDR4 back from the dead. A decade old AMD AM4 platform surging 40% in popularity. Five year old Intel LGA 1700 seeing renewed interest. Micron Virginia expansion quadrupling DDR4 wafer output. Spot prices up 70%. There is no escape.

Open on X

@itsmichaelluuJun 4, 2026 · 3:09 PM

$DRAM added 2 out of 100 calls. The idea is simple: Buy $DRAM when it's cheap, hold until my target of $200+ possibly $300. Here's what $DRAM ETF holds: 🔹 SK Hynix (~26%) — World's #1 HBM memory supplier to AI data centers 🔹 Samsung (~22%) — Memory giant powering servers, smartphones, and AI infrastructure 🔹 $MU (~25% combined exposure) — America's memory champion benefiting from AI demand 🔹 Kioxia — Leading NAND flash storage manufacturer 🔹 $SNDK — Consumer & enterprise storage solutions 🔹 Seagate — Hard drives and AI data storage 🔹 Western Digital — Storage infrastructure for hyperscalers Why I like it: ✅ AI factories need memory ✅ HBM demand is exploding ✅ Supply remains constrained ✅ Memory pricing is rising ✅ One ETF gives exposure to the entire memory cycle My thesis: Every AI factory needs GPUs. Every GPU needs memory. Memory is becoming one of the biggest bottlenecks in AI. $DRAM = one of the simplest ways to invest in that trend. ♻️ RESHARE this post and write 1 comment, if you want to follow along and get my entries too.

Open on X

@tradexwhispererJun 3, 2026 · 5:55 PM

$MU $SNDK Unlimited $DRAM I know it sounds crazy But the more I think about it, the scale of memory required to deploy AI Agents across every enterprise and every household is almost incomprehensible. Think about it this way. To give every human on the planet a capable AI Agent, we need to virtually replicate the equivalent of a biological brain for each person. Every neuron maps to a transistor. Every memory trace maps to stored bits. 86 billion neurons. 8 billion people. Do the math. The number is not large. It is effectively infinite from a productivity and demand standpoint. We are not building a product. We are building a second layer of human cognition at planetary scale. The demand is not massive. It is effectively limitless.

Open on X

+ 69 more

$DRAM

$DRAM

About this stock

Related stocks

Frequently asked