Market Brief(X) — May 25–May 28, 2026

2026-05-29 • Twitter

Executive Summary

The AI infrastructure supercycle accelerated this week with MU crossing $1T market cap on UBS’s valuation re-rating to $1,625, fueled by HBM/LTA structural repricing that rewrites memory from cyclical to growth-equity multiples. The week’s most important structural signal was the broadening of AI constraints from GPU/HBM into passives (MLCC shortages spreading from high-end to general), power semis (Infineon’s second price hike in three months), and PCB equipment, confirming that AI bottlenecks are cascading down the supply chain into physical layer components with asymmetric pricing power. Mid-week, SNOW’s +35% post-earnings move validated the software-AI monetization thesis that had been dismissed, triggering a rotation signal from hardware into select software names. The central tension: sentiment is euphoric at the single-stock level (QCOM IV Rank 100, MRVL pre-earnings call skew, retail FOMO into optics), yet the macro liquidity backdrop is tightening into month-end pension rebalancing and a crowded PCE print that came in slightly below expectations — enough to sustain the bid but not enough to justify chasing extended positions.

Key Themes & Trends

Memory’s Structural Re-rating: From Cyclical to Growth Equity

The dominant theme of the window. UBS lifted MU target to $1,625, anchoring on FY29 trough EPS of $117 × 15x PE — a deliberate shift from PB to PE valuation methodology driven by long-term agreement (LTA) penetration that smooths earnings volatility (@RichTerry123). SK Hynix also crossed $1T market cap mid-week (@jukan05). The Korean press revealed that Alphabet, Microsoft, and Meta offered to fund SK Hynix’s Yongin fab construction, but SK Hynix declined — instead using the desperation as leverage for more favorable LTA terms with prepayments, 5-year+ commitments, and price-floor guarantees (@jukan05). This “super-eul” (super-supplier) dynamic is the core regime change: memory makers are converting customer desperation into contract structures that lock in high-margin revenue visibility for years. @qinbafrank connected this to SanDisk’s new business model with collateralized LTAs. The key insight from @ivanalog_com: memory supply is effectively locked — there is no fourth Western supplier, making the competitive structure closer to GPU oligopoly than traditional DRAM. Dissent came from @TJ_Research, who noted that UBS’s own EPS projections (155 → 167 → 117) prove cyclicality still exists, undermining the re-rating thesis itself. MU EVP confirmed meaningful supply only comes online in late 2027, ramping in 2028 (@jukan05).

Mid-window arc: The theme was already in motion Tuesday with the UBS call, then confirmed Wednesday with SK Hynix’s $1T milestone and the Korean fab-rejection story, and extended Thursday into broader semis via DELL’s blowout AI server numbers (Q1 AI server revenue $16.1B, +757% YoY) and raised full-year AI server guidance to $60B (@qinbafrank).

MLCC & Passive Components: The Next Choke Point

This theme emerged strongly mid-window (Wednesday) and accelerated through Thursday. @jukan05 reported that China’s largest passive component maker Fenghua suspended order intake for 0402/0603 MLCCs and resistors, with utilization at ~90%. Murata and Samsung Electro-Mechanics utilization surpassed 90%, with Samsung’s Tianjin plant “flat out.” Lead times stretching to 12-18 weeks; Yageo signaled price hikes as book-to-bill hit 1.3 (AI products at 1.4) (@jukan05). @AntonLaVay produced a deep-dive explaining the structural nature: AI’s power delivery network (PDN) requires high-capacitance, low-ESL, high-voltage MLCCs near the GPU/ASIC — a supply-constrained, certified, high-spec market where the component is low-cost to the rack but catastrophic to lose, creating “asymmetric pricing power.” The shortage is spreading from high-end AI to standard MLCCs as Japanese/Korean capacity pivots to AI, squeezing general supply — a “weakened version” of what happened with memory (@AntonLaVay). @LinQingV called out Musashi (7220.T) as the sole global supplier of hybrid supercapacitors for NVIDIA’s GB300 power architecture, with a visible supply-demand gap through 2027. @zephyr_z9 flagged that he will publish an MLCC equipment and raw material supplier list, distinguishing the picks-and-shovels from the manufacturers themselves.

Profile mix: Industry analysts (Jukan, Zephyr, LinQing) + trader-investor (AntonLaVay, labubu_trader) + macro commentator (fin) — diverse convergence strengthens the signal. @jdhasoptions explicitly shifted portfolio toward MLCC and power semi, trimming NVTS/WOLF (too volatile) for Japanese equity exposure.

Software AI Monetization Inflection: SNOW Breaks the Dam

The most significant contrarian signal of the week. SNOW reported Q1 revenue +33% YoY, RPO +38%, and announced a $60B 5-year AWS AI compute commitment — the stock surged ~35% after hours Wednesday (@ArtofSpecuycky). The market interpreted the AWS commitment as Snowflake putting real capital behind AI workload expectations they can already see in their pipeline. @FundaAI had been flagging Cortex Code as “the fastest-growing product in Snowflake’s history” since their Q1 preview, calling it correctly across three consecutive quarters. The broader implication: AI monetization is moving from hardware into the software/data layer, validating the rotation thesis. MDB followed Thursday after hours with strong results (RPO +88%, cRPO +69%), confirming the pattern (@ArtofSpecuycky). @ShanghaoJin called this “the first firework” in the thesis that “short IGV long semi” would break. Market participants began reassessing which software names are AI beneficiaries vs. victims (@NullOreo_ cited DT, ESTC, RBRK, VRNS). @qinbafrank argued vertical SaaS with proprietary data + workflow + compliance moats would prove more durable than horizontal tools.

Within-window: SNOW Wednesday after hours → MDB Thursday after hours → IGV broke above 200-day MA Thursday, triggering the rotation signal. The arc validates earlier calls and shifts the conversation.

CPO/Optics: Euphoria Meets Execution Reality

The optics/CPO complex ran hot early in the window but faced increasing skepticism about timeline and execution risk. Rosenblatt checks revealed NVIDIA asked supply chain for ~20x InP laser capacity increase from 2025-2030; vendors conservatively agreed to ~12x (@zephyr_z9). @ShanghaoJin pushed back hard against the narrative of rapid CPO scaling, arguing InP fab capacity cannot expand fast enough — MOCVD tool lead times are 18 months, epi yields are inherently low, and the talent constraint is real. He ranked the 2028 capacity hierarchy: NOK, LITE, Cohr at top tier, Chinese players next, AAOI in third tier — and questioned SIVE’s capacity sourcing entirely. @Corsica267 noted optics names did not bounce with the broader market on Thursday, hypothesizing they may be more rate-sensitive than understood, or that the market is “pricing the timeline too optimistically, completely ignoring yield and capacity ramp.” He recommended waiting for The Information or similar to report on switch redesigns or CPO yield issues before buying the Serenity-recommended small-cap names. The nuance: CPO as a direction is “certain,” but the path is “inevitably bumpy,” and the market is pricing it as linear (@ShanghaoJin, quoting the I Ching: “Wind on the mountain — gradual, enduring, not hurried”).

AI Infrastructure Diffusion: Power, PCB, Capacitors, Silicon Capacitors

The theme of “what’s next after GPU/HBM” expanded into multiple sub-themes. @jukan05 reported Intel’s EMIB packaging is driving silicon capacitor adoption, with Google’s v8e chip as the first confirmed application. Samsung Electro-Mechanics (SEMCO) landed a >$1B silicon capacitor supply contract for Intel EMIB from effectively zero revenue last year — the world’s only company operating both silicon capacitor and substrate businesses, giving it turnkey leverage (@jukan05). @LinQingV published a detailed analysis of China’s PCB equipment capex cycle: 9 leading PCB companies’ 2025 total capex of ¥26.7B (+111% YoY), Q1 2026 at ¥12.5B (+182% YoY), driving demand for laser drilling (Han’s CNC), LDI exposure (CFMEE), and VCP electroplating (Dongwei) — upstream equipment suppliers benefiting from downstream homogenized capacity expansion. @LinQingV also covered ByteDance’s in-house CPU development evaluating both Arm and RISC-V, contextualizing it within the Agentic AI era where CPU is the “orchestration layer” for thousands of concurrent agent sessions, making server CPU a critical bottleneck.

Geopolitical Whiplash: Iran Deal “Will They/Won’t They”

The US-Iran narrative oscillated wildly within the window: Tuesday saw a rumor of an Iran-proposed memorandum with 30-day Strait of Hormuz mine clearance, causing oil to dip and AI to take profits into airlines/cruises. The White House quickly denied it (@qinbafrank). Thursday brought a US-proposed version, which Iran denied. US airstrikes resumed mid-week. @qinbafrank characterized the dynamic as “the ball is in the penalty area, but there’s a scrum at the goal mouth” — both sides want a deal but are fighting over who gets to claim victory and on what terms. The market has been conditioned to buy every dip on peace rumors, but the tail risk of a breakdown remains the single largest macro variable.

Market Sentiment

Short-term (next 1-2 weeks)

Bullish with caution. @ArtofSpecuycky sees the path as: continued strength into ~June 12 (SpaceX IPO), a 3-5% pullback June 15-18 (FOMC + OpEx), then resumption higher into late July/early August, with a larger correction around midterms. @KevinXInvest sees a local top at 7575-7590, 250pt dip, then push to 7700. @Corsica267 called for caution May 28-June 2 due to liquidity tightening, advising to trim trading positions and add rate-sensitive core — this was validated as the week ended with technical overbought signals accumulating (MACD/RSI negative divergence at ATHs, 9-week winning streak historically sees pullbacks within 3 days in 80% of cases). @mat78704 provided the short-term tactical read: buying opportunities Thursday/Friday, rally expected next week into June 3 highs, with SOXX looking weak suggesting sector rotation.

Long-term (weeks to months)

Structurally bullish but with growing acknowledgment of concentration risk and valuation stretch. @RichTerry123 captured the structural tension: AI upstream is booming, but the downstream monetization question remains unanswered — “if downstream consumption can’t keep up, upstream will eventually shoot itself in the foot.” @ShanghaoJin framed the macro as “deregulation releasing liquidity nonlinearly,” predicting a “low-PE bubble” that is different from 2000 — “this time there’s genuine demand and genuine shortages” — but whose “Achilles’ heel is Anthropic.” @ivanalog_com offered the most provocative endgame scenario: “Ten largest market caps, seven are semiconductor companies, the bottom three MAG7 (the biggest AI CAPEX spenders) drop out, their financials deteriorate, internal anti-investment factions purge the spenders, CAPEX collapses — then Phase 1 AI correction begins.” This represents an important dissenting framework within an otherwise bullish consensus.

Sentiment Balance & Shift

The week saw extreme divergence between price action (ATHs) and sentiment indicators. AAII bearish sentiment spiked to 43.6%, above historical mean of 31%, while bullish fell to 31.7% (@ArtofSpecuycky). Short interest on S&P 500 median stock hit 3%, highest since 2012, 2x the COVID crash level — this is squeeze fuel, not distribution (@ArtofSpecuycky). Fear & Greed at 71, not yet at Extreme Greed (75+). The striking feature: index at ATHs while retail is increasingly bearish, institutions are heavily short, and breadth is narrow — this is not a typical euphoric top. However, single-stock euphoria is unmistakable: MU options showed massive deep-OTM call buying, QCOM IV Rank at 100, MRVL pre-earnings call skew was extreme. The split between macro caution (liquidity tightening, technical overbought) and sector-level euphoria (memory, MLCC, space) is the defining sentiment tension.

Key Figures & Assets

Short-term / Technical Trades

  • $IREN: @ArtofSpecuycky highlighted a $4.6M single-leg call buy for Aug 70 strike, OTM 16.2%, purchased at 11am on news of the Dell $1.6B procurement deal. First target achieved at 65, next target 77. He expects a potential SW1 contract announcement with Anthropic around June 8 as the next catalyst. @ArtofSpecuycky maintained his view from the 48-50 buy zone (+37% since), with the key level to watch being 60 holding.
  • $QCOM: @KotlinerBTC noted IV Rank at 100 with no earnings effect — extremely unusual and suggests “something big is coming.” Put/Call skew deeply negative, call accumulation heavy above 250. @tradergokux flagged a volume-contraction pullback to the 5-day EMA as a setup; QCOM+ByteDance ASIC deal and a rumored additional US hyperscaler project (Amazon, per @jukan05) are the fundamental catalysts. @ArtofSpecuycky noted forward PE ~23x, PEG 1.06 — “not expensive for what it is.”
  • $AOSL: @tradergokux shared the gap-up (+11%) as a continuation of a setup shared earlier; breakout from consolidation.
  • $ENPH: @tradergokux identified a textbook cup-and-handle breakout, shared with his group.
  • $ASTS: @labubu_trader looking to buy the dip to 5d EMA (~119) or 10d EMA (~107) after Blue Origin’s New Glenn static fire explosion, viewing ASTS as an indirect beneficiary of competitor delays. @KotlinerBTC noted that after three consecutive up days, the call-buying bid has returned and IV is breaking out, suggesting the previous “bulls exiting ahead of SpaceX IPO” concern may be resolving.
  • $PLTR: @KotlinerBTC observed a contrarian signal: despite the stock falling -2.99%, short-dated 140-150 calls saw heavy buying Wednesday, with 86-day 140C volume spiking. Put/Call skew moved toward calls on a down day — “someone knows something.” This setup conflicts with PLTR’s YTD -23% performance, making it a high-conviction divergence signal.

Long-term / Fundamental Positions

  • $MU / Memory complex: The re-rating thesis is detailed above. @jukan05 relayed management’s confirmation that new investment adds only ~15% bit volume; HBM consumes so much DRAM bit volume that actual DRAM product bit growth is near zero. The supply constraint extends through at least 2028. TrendForce projects global DRAM revenue +303% YoY in 2026, +46% in 2027 (@jukan05). Risk: @jukan05 noted that hyperscalers are reluctant to renegotiate HBM prices signed under last year’s LTAs — the pricing power may have limits.
  • $NOK: Convergent conviction across multiple profile types. @AntonLaVay highlighted seven insider purchases over the last 30 days, totaling $2.5M+, with CEO Justin Hotard buying €772K at €9.15. He noted the stock is “controlled and disciplined” — trending along EMA10 with EMA10/20 maintaining separation (@AntonLaVay). @ShanghaoJin framed NOK as a top-tier InP fab capacity holder through 2028 alongside LITE and Cohr. @ArtofSpecuycky called NOK one of his two highest-conviction names alongside MRVL.
  • $ORCL: @KotlinerBTC observed IV beginning to move despite flat price — “volatility is the first to move.” Expected earnings June 10; IV trend suggests 100%+ volatility at earnings. @ShanghaoJin took a position on the breakout through the call wall. @AntonLaVay connected the dots: “NBIS and CRWV are moving well — really not looking at ORCL?”
  • $BB: @qinbafrank published a detailed thesis: BlackBerry’s QNX is positioned as the safety-critical OS layer for Physical AI (robotics, autonomous vehicles, medical). QNX Q4 revenue +20% YoY, backlogged royalty orders at $950M pre-locked. The key catalyst: QNX OS for Safety 8.0 integration with NVIDIA IGX Thor, extending beyond automotive into robotics and medical. @ArtofSpecuycky added technical confirmation: broke above a 14-year resistance at 6.3-6.5, opening upside.
  • $INTC: @jukan05 reported Intel is investing “several trillion won” in EMIB packaging expansion, with orders being placed now, full-scale operation expected next year. The foundry revival thesis through packaging differentiation rather than front-end process competition is gaining traction. Samsung’s Taylor fab confirmed “ready for operation” with mass production next year, focusing on 2nm for Tesla AI5/AI6 and others (@jukan05). The new Intel narrative: EMIB + 18A + packaging may be the start of a genuine foundry comeback.

Trading Activity & Holdings (VIP & High-Weight Traders)

  • @labubu_trader (High): Holds NOK (“very big position since $7.5”) (source). Added QCOM 2X leveraged at $214.5 on Thursday night market (source). MLCC basket priority: Taiyo Yuden > Musashi > Murata = Yageo > others; thinks Taiyo and Musashi both have 2x+ potential (source). SNOW position generated $200K profit on the post-earnings gap-up, holding through $230 level (source). Traded WOLF in the prior two weeks, calling it “good, just too much volatility” (source). Also holds airlines + NKE + F/CLF calls (source).
  • @KevinXInvest (High): Hedged long-term position at SPX 7530 with SPX Oct 7000/6500 puts and SPY Oct 700 puts; also sold calls (source, source).
  • @ArtofSpecuycky (VIP): Heavy IREN position, added at 48-50 zone, scaling sell orders at 77. MRVL long-term holder. NOK and MRVL remain highest-conviction names. Small RKLB position — plans to exit entirely before June 12 SpaceX IPO (source). ALAB core position. SNOW position relatively small, would consider adding on pullback. RDW bought small position of options for fun (source).
  • @AntonLaVay (Medium, but high signal on MLCC): Holds 6981 (Murata) and 6976 (Taiyo Yuden); considers himself primarily a Japan equity investor by capital allocation (source). NOK held via 2027 LEAPS (source). A-share MLCC proxy via Fenghua Advanced Technology for his mother’s account (source).
  • @Corsica267 (High): CIEN average cost at 213 (source). Trimmed half of optics positions Thursday (source).
  • @JD (VIP): Shifting portfolio toward MLCC and power semi; trimmed NVTS/WOLF for Japanese equity exposure (source). Now position from 104 cost basis (source). VSH added (source). Noted selling SNDU for INTC, which he found surprising (source).
  • @Franktradinglog (High): NOK (“好人啊，买啊”) (source). Optics complex via LITE, AAOI, GLW, with AAOI 0DTE puts as a hedge (source). LLY, F, FSLR, NXT positions generating significant returns (source).

Off-Theme Highlights

  • $FSLR: @NullOreo_ celebrated a significant move — “no one noticed FSLR earlier than me.” The solar trade is running independently of the AI infrastructure theme.
  • $WOLF: @NullOreo_ argued WOLF’s unique value as “the only complete SiC vertical supply chain in the US” — a national security angle not yet priced in.
  • $XPEV: @ivanalog_com highlighted XPEV’s Q2 delivery guidance remaining flat YoY despite subsidy headwinds, a standout among Chinese EV peers. Also flagged the underappreciated robotics division — “a rare low-valuation humanoid robotics company.”

Notable Perspectives & Insights

@RichTerry123: “The market isn’t buying companies, it’s buying stories”

A sharp critique: HPQ reported poor results (memory costs compressing PC margins), yet DELL and HPE surged on “AI server” labels. IONQ (sub-$50M revenue) and RKLB (still losing money) are pricing in the most optimistic decade-out outcomes. The distinction: “叙事加分，盈利不加分，一个季度 miss 直接出局” (Narrative scores points, earnings don’t — one quarterly miss and you’re out). This is not a prediction of imminent decline but a warning that those chasing these names must understand they’re buying “the duration of sentiment, not company value” (@RichTerry123).

@ShanghaoJin: “Low-PE Bubble” and the Anthropic Achilles’ Heel

Running a fever and “thinking in a daze,” he sketched the framework for how this cycle breaks: genuine shortages → earnings growth → PE compression → stock rallies → continued price hikes. The cycle can run for two years, even through 2028 “with no cyclicality.” But “the market cannot keep turning like this indefinitely.” The trigger: Anthropic. “I just haven’t figured out the fuse yet.” The logic: Anthropic revenue deceleration would be the first crack in the “AI demand is infinite” narrative that justifies all upstream CAPEX. When it comes, “one game, one fake news story could collapse the whole thing” (@ShanghaoJin). This is arguably the most important contrarian framework in the entire feed — a specific, falsifiable thesis about where the cycle breaks.

@NullOreo_: The Liquidity Math on MU

A concrete quantification: MU at ~$1T market cap, UBS target $1,625 implies ~$1.85T. To get there requires finding $100-300B in incremental buyer capital. “Does the market currently have the funding capacity, and does that capital have the willingness to take on that exposure? This needs specific research.” This frames the bull case in practical liquidity terms rather than narrative (@NullOreo_).

@jukan05: The HBM Renegotiation Problem

“Hyperscalers appear very reluctant to renegotiate the prices under the long-term agreements they signed last year.” The implication: HBM has shifted from spot-market DRAM commodity to “customer-bound, technology-bound, capacity-bound quasi-custom memory” — higher certainty but can no longer fully capture spot-price repricing in an upcycle. Ironically, Micron’s relative weakness in HBM may become a good thing due to soaring traditional DRAM pricing (@jukan05). This is a subtle but critical point about the limits of memory re-rating.

@Corsica267: From Analyst to State Machine

A deeply personal reflection on trading evolution. The core insight: “Identify states, not predict the future. Reduce prediction, reduce personal interpretation. Be a state machine.” He’s abandoned macro analysis — rate curves, dollar, fiscal flows — in favor of a cross-asset state detection system that’s been live for a month, losing only “$10.” The cost: trading has become boring, “like clocking into work.” This matters because it reflects a broader shift among sophisticated participants away from narrative-driven positioning toward systematic, state-dependent execution — reducing the influence of “analyst” commentary on actual capital allocation (@Corsica267).

The Read

The week was defined by the tension between the undeniable strength of the AI infrastructure buildout and the increasingly narrow foundation on which the rally rests. On one side, the data is extraordinary: MU crossing $1T on a genuine re-rating thesis, DELL reporting +757% AI server growth and guiding to $60B, SK Hynix rejecting Big Tech’s fab funding because business is too good, MLCC makers suspending orders because they literally cannot produce enough, and SNOW proving that AI software monetization is real. These are not narratives — these are orders, shipments, and prices clearing at higher levels. The “this time is different” argument for memory’s structural transformation has more evidence behind it than any prior cycle, and the diffusion into passives, power, PCB, and equipment is classic late-cycle breadth expansion that typically extends trends rather than ending them.

On the other side, the cautionary signals are piling up at the tactical level. Technical divergences at index highs (MACD/RSI negative divergence, VIX divergence), extreme single-stock sentiment readings (QCOM IV Rank 100, MRVL pre-earnings call skew, retail flooding into CPO names on margin), and a macro liquidity backdrop that is quietly tightening into month-end with pension rebalancing adding selling pressure. The market is simultaneously pricing peak euphoria in the hottest names and peak skepticism in the index, as measured by AAII bearish sentiment above 43% and median stock short interest at decade highs. This is a squeeze setup, not a distribution setup — but squeeze dynamics work both ways, and the unwind, when it comes, will be sharp.

The most important development this week was not the MU price target or the DELL blowout, but the broadening of the AI constraint set into components (MLCC, silicon capacitors, supercapacitors) and equipment (PCB, packaging, test). This confirms that the AI buildout is now constrained at the physical layer — not just at the leading-edge semiconductor node, but at the mundane, low-ASP, high-volume, hard-to-substitute components that every rack needs. This is where the asymmetric pricing power lives, and it’s where the next leg of alpha will be generated. The commentators who recognized this early — @AntonLaVay on MLCC structure, @LinQingV on PCB equipment, @jukan05 on silicon capacitors — are pointing toward a playbook that is less crowded than the memory/optics trade.

The risk: the window is seeing a classic “good news is priced, great news is needed” dynamic. MRVL beat and raised, and the stock initially sold off on the news because expectations had run so far ahead. The same dynamic awaits AVGO next week. When expectations are this elevated, the bar for positive surprises keeps rising, and the penalty for mere in-line results increases.

What to Watch

  • PCE Aftermath & Month-End Flows (Fri-Mon): April core PCE came in at +0.2% M/M vs. +0.3% expected — slightly dovish (@kayliatyyy). But the market now needs to digest this alongside month-end pension rebalancing (~$20B equity selling expected), which could create choppy conditions into month-end. Watch whether the soft print is enough to sustain the bid or if liquidity effects dominate.

  • SpaceX IPO Pricing & Flow Dynamics (June 12): SPCX reportedly lowered its targeted IPO valuation to “at least $1.8 trillion” from earlier higher expectations (@Balder13946731). The lower price could be a positive surprise or a sign of weaker-than-expected demand. The key dynamic: @ArtofSpecuycky expects a liquidity-siphoning effect on the IPO date, while @qinbafrank plans to sell RKLB into SpaceX-driven enthusiasm rather than hold through the event. RKLB’s $3B ATM equity raise adds supply overhang to the space sector.

  • AVGO Earnings (June 3): The next major AI infrastructure read. With MRVL having set a high bar and the stock initially selling off despite beating, AVGO faces a similar dynamic. @KotlinerBTC noted AVGO IV rising sharply, linked to MRVL’s earnings effect spilling over. The market will be looking for custom ASIC ramp confirmation and any updates on the hyperscaler ASIC pipeline (especially the rumored MTK/xAI/SpaceX ASIC win on Intel 14A + EMIB-T, per @jukan05).

  • FOMC + OpEx (June 18): @ArtofSpecuycky identified this as the highest-probability window for a 3-5% pullback. Warsh’s first FOMC meeting as Chair, combined with quarterly OpEx, creates a volatility event. The market’s pricing of a potential 2027 rate hike is a tail risk — any hawkish shift in the dot plot or commentary would hit the long-duration AI growth trade.

  • MLCC Supply-Chain Data Points: Monitor for further order suspensions, price hike announcements, and lead-time extensions from Murata, SEMCO, Taiyo Yuden, and Yageo. @zephyr_z9 will publish his MLCC equipment/raw material supplier list — this could catalyze the next leg of the MLCC trade beyond the component makers themselves.

  • Corsica267 Liquidity Framework: @Corsica267 called for caution May 28-June 2 based on liquidity tightening signals from his proprietary state-machine system. His framework flagged that rate-sensitive sectors should be added as core positions, with cash equivalents shorted. With this window now closing, watch whether he signals a re-entry point into the week of June 2, which would align with the post-month-end, pre-SpaceX window that multiple commentators have identified as constructive.