[Research Brief] May 29, 2026 — Semis Supercycle Debate, AI Infrastructure Specializes, Bitcoin Drifts Through "Impasse"

🧭 MACRO SNAPSHOT

Risk assets are printing record highs (S&P 500 at 7,520; VIX subdued at 16.29) on a combination of Iran de-escalation hopes and continued AI-driven mega-cap leadership (MarketWatch). The Fed Funds rate sits at 3.64% with the 10Y at 4.48% and 2Y at 4.0% — a normalized but still positive curve, with HY credit spreads compressed to 271 bps signaling no stress in risk markets (FRED). GDP growth at 1.6% with unemployment at 4.3% is consistent with a soft-landing-but-decelerating regime; today's PCE inflation print is the key event risk for rates positioning.

Two macro currents matter for the portfolio: (1) semiconductor "supercycle" framing is becoming consensus (MarketWatch is now openly debating whether to treat chips like commodities — typically a late-cycle signal worth respecting given our overweight in $NVDA, $AVGO, $TSM, $MU, $AMD); and (2) consumer credit stress is building beneath the surface — CNBC reports >$1,000 monthly auto loan payments are now common on non-luxury vehicles, which is relevant to $AFRM's underwriting and $TSLA's affordability narrative. Noahpinion's "debtpocalypse" piece (newsletter signal) reinforces the long-duration risk premium that has kept 10Y yields elevated.

₿ BITCOIN DAILY WRITE-UP

Price & Market Structure
Bitcoin sits at $73,333, flat on the day, -5.16% on the week and -5.49% on the month. BTC remains ~42% below the $126K ATH printed earlier this cycle, and dominance has climbed to 57.7% — meaning BTC is outperforming altcoins on the way down, a classic risk-off crypto signal. This is the third consecutive month of flat-to-down price action 13 months post-halving — structurally divergent from prior cycles.

Structural Thesis
The core reason to hold remains intact: spot ETFs are a permanent demand sink, the post-halving supply compression is mechanical and irreversible, and institutional/sovereign accumulation has structurally raised the floor. We own Bitcoin as a long-duration call option on monetary debasement and digital reserve asset adoption — not as a trade.

What Happened This Week
The most notable structural development is Fidelity's framing (via Bitcoin Magazine) that miners are facing an "AI squeeze" as hash rate flattens and the network enters a new security phase — miners are increasingly tempted to repurpose energy/compute capacity for AI workloads, which has real implications for hash rate growth. Separately, The Bitcoin Layer published "Impasse" arguing we are in the late stage of a bear market — notable because TBL is generally structurally bullish. No material ETF flow data this week, which remains a flagged data gap.

Bull / Bear Scorecard

Bull:
- Halving supply shock still mechanically working through; ETF demand persists
- Institutional plumbing (custody, ETFs, treasury frameworks) is infrastructural and non-reversible
- Sovereign accumulation continues to broaden the buyer base

Bear:
- 38.8% drawdown 13 months post-halving is historically anomalous; cycle peak at $126K cannot be ruled out
- TBL's "Impasse" framing — a typically bullish analyst calling late bear — is a sentiment marker worth respecting
- Miner economics deteriorating as AI competes for energy/capex — could pressure hash rate and network security narrative

Conviction Check: Action: HOLD | Conviction: 7/10. No change. The price action is uncomfortable but the structural thesis is not broken; what would break it is sustained ETF outflows or a confirmed LTH distribution wave — neither confirmed yet.

What to Watch
- Spot BTC ETF net flows (rolling 30-day) — sustained outflows >30 days = thesis break
- Glassnode LTH supply metric — rotation back to accumulation = cycle-extending signal
- Real yields (DFII10) and DXY — a break lower in either is the macro unlock

Community Pulse
Reddit was completely silent today across all tracked subs — unusual for a Friday and itself a sentiment data point (apathy, not panic). Newsletter sentiment has turned cautious: The Bitcoin Layer's "Impasse" and "Everything Everywhere All at Once" posts both lean toward late-bear framing, and Bitcoin Magazine's miner squeeze piece adds a structural concern that wasn't on the radar 6 months ago. The dominant tone is patient resignation — not capitulation, but the parabolic post-halving narrative is being quietly retired.

🔬 TODAY'S DEEP DIVES

SMAR — Smartsheet Inc. — NEW IDEA (SCREENED, NOT ADDED)
Conviction: 4/10 | Status: WATCHLIST | Sector: Technology (Software)

WHAT THEY DO: Smartsheet is a cloud-based collaborative work management (CWM) platform — think project tracking, workflow automation, and team coordination built around a spreadsheet-like interface. They make money via per-seat SaaS subscriptions, primarily to enterprise customers who need structured collaboration across functions (marketing ops, IT, construction, etc.).

WHY IT'S INTERESTING NOW: Honestly, it isn't — and the data feed told us as much by returning null across every field. The most likely explanation is that Smartsheet was taken private (Vista/Blackstone consortium reportedly closed a deal). If still public, the AI-driven workflow automation thesis would be relevant, but Microsoft Copilot + Loop is a serious bundle threat.

BULL CASE:
- AI-driven workflow automation expands TAM if Smartsheet captures share via enterprise governance features
- Enterprise CWM consolidation favors integrated platforms over point tools
- FCF margin expansion toward 25%+ as S&M efficiency improves
- M&A interest provides a valuation floor

BEAR CASE:
- Most likely outcome: stock no longer publicly traded — analysis moot
- Microsoft Copilot + Excel/Loop bundling compresses SMAR's value prop
- Mid-market churn risk in recession; net retention deteriorates

KEY METRICS: Unavailable — entire data feed null.

BOTTOM LINE: Not added to target list. 4/10 conviction reflects fundamental data unavailability, not a negative view — we need to confirm public status before this gets another look.

MP — MP Materials Corp. — ROLLING REVIEW
Conviction: 7/10 (reaffirmed) | Status: WATCHLIST + Portfolio Hold | Sector: Materials

WHAT THEY DO: MP operates Mountain Pass in California — the only scaled rare-earth mining and processing operation in the Western Hemisphere. They sell rare earth concentrates and separated NdPr oxides to magnet producers, and increasingly sell finished NdFeB magnets directly to OEMs. Anchor customers: General Motors (long-term EV traction motor supply agreement, 2021) and the U.S. Department of Defense (direct equity stake, 2025).

WHY IT'S INTERESTING NOW: Two converging catalysts: (1) the DoD equity investment essentially makes MP a strategic national security asset with implicit pricing/offtake support, and (2) the EV magnet supply chain is bifurcating between China and the West, and MP is the only credible Western pure-play. Stock is +241% over 1Y at $66.83 — the re-rating has been violent, so this is now a "ride the structural thesis but acknowledge valuation risk" position.

BULL CASE (synthesized from prior thesis):
- Only scaled Western rare-earth processor — irreplaceable in any China-decoupling scenario
- DoD as anchor equity holder provides implicit demand and pricing floor
- Magnet-finished product expansion lifts margins materially vs. selling raw concentrate
- Secular EV + defense + wind tailwind drives NdPr demand growth >2x by 2030

BEAR CASE (synthesized from prior thesis and market context):
- After +241% in 1Y, valuation has run ahead of near-term fundamentals
- Rare earth prices are still set by Chinese producers; MP is a price-taker globally
- Magnet manufacturing ramp execution risk — vertical integration is harder than mining
- If China relaxes export restrictions tactically, prices crater

KEY METRICS: 1Y return +241%; market cap reflects strategic premium; gross margin trajectory depends on magnet-finished mix.

BOTTOM LINE: Maintain 7/10. Holding in portfolio with selective trim opportunity — the thesis is intact but the easy money has been made.

📋 TARGET LIST STATUS

Conviction changes this week: Five reaffirmations on rolling 50-day reviews (UUUU, FCX, ASTS, RKLB, AVAV) — no upgrades, no downgrades. No names dropped today. SMAR did not earn a spot (4/10 conviction + null data feed). The list remains at 47 names with room for 3 high-conviction additions.

💼 YOUR PORTFOLIO

• $AAPL | HOLD | 5/10 — All-time highs on Bofa $380 target raise and Globalstar/Amazon transaction; price has outrun fundamentals, conviction lowered to reflect valuation discipline.
• $AVAV | HOLD | 7/10 — Class action lawsuits are noise; durable position in attritable UAVs/defense autonomy remains intact.
• $AVGO | STRONG HOLD | 8/10 — Custom silicon + VMware moat; Wi-Fi 8 SoC launch reinforces the diversified non-NVDA AI infrastructure thesis.
• $BABA | STRONG HOLD | 8/10 — Down 9.9% in 3 weeks, -34.5% from 52W high; the gap between cloud/AI fundamentals and stock price is widening — this is where conviction matters.
• $FSLR | TRIM | 6/10 — +59% in a month is a gift, not a thesis improvement; trim into strength.
• $GOOGL | STRONG HOLD | 9/10 — Gemini 3.5 + Spark agent at I/O 2026 confirms platform leadership; the cheapest mega-cap on a fundamentals-per-dollar basis.
• $ISRG | BUY MORE | 8/10 — Stock at 52W low, -23% on the year; this is the dislocation we wait for in compounder names.
• $MKL | HOLD | 7/10 — Quiet quarter, +103% unrealized — let it compound.
• $MP | HOLD | 7/10 — See deep dive; ride the thesis, watch the valuation.
• $SYM | HOLD | 6/10 — -14.9% MoM; no material news, monitoring for fundamental confirmation.
• $TSLA | TRIM | 4/10 — SpaceX stake/IPO optionality is interesting but doesn't justify current valuation given core auto deceleration; trim continues.
• $UNH | STRONG HOLD | 7/10 — Recovered to $382 from $234 low; PBM transparency push reduces tail risk, Bernstein Outperform reiterated.

⚠️ WATCH LIST

• $MU — +78% in a month, now in the $1T club. Conviction at 6/10 likely too low if HBM pricing holds; will upgrade on confirmation of HBM4 pricing discipline and FY27 guidance.
• $LITE — Down 10.77% on the week despite intact AI optics ramp. Watching for any sign the transceiver order book is slowing; absent that, this is a buying opportunity, not a downgrade.
• $PDD — Down 15% on the week post-earnings miss; conviction 7/10 will be retested if Shein litigation escalates or if take-rates compress further.
• $ENPH — +122% in a month with one news item (PowerBridge patent acquisition). This kind of move without fundamental catalyst is a red flag — monitoring for thesis confirmation or fade.

🔁 RE-REVIEW QUEUE

• $CEG | Was 6/10 | Dropped 2026-05-13 — Dropped on conviction threshold. Power demand from AI data centers (xAI Colossus 2 at gigawatt scale, per SemiAnalysis) materially strengthens the nuclear/IPP thesis; worth a fresh look.
• $DDOG | Was 6/10 | Dropped 2026-05-07 — Dropped on conviction threshold. AI observability TAM expansion (per SNS Insider report on $182B monitoring tools market by 2035) may have improved the setup; worth re-screening.
• $TTD | Was 6/10 | Dropped 2026-05-07 — Dropped on conviction threshold. CTV ad market dynamics unchanged; lower priority for re-dive.
• $BWXT | Was 6/10 | Dropped 2026-05-13 — Dropped on conviction threshold. Small modular reactor narrative compounding with data center power demand; warrants a fresh look.
• $NET | Was 6/10 | Dropped 2026-05-07 — Dropped on conviction threshold. Edge inference thesis (Rubin CPX prefill specialization) is increasingly relevant to Cloudflare's positioning; worth re-screening.
• $KTOS | Was 6/10 | Dropped 2026-05-13 — Dropped on conviction threshold. Defense autonomy tailwinds intact (parallel to AVAV); lower priority given existing AVAV exposure.
• $CRWD | Was 6/10 | Dropped 2026-05-08 — Dropped on conviction threshold. Cybersecurity sector running hot ($PANW +42% MoM); worth re-screening for relative value.
• $NTRA | Was 6/10 | Dropped 2026-05-09 — Dropped on conviction threshold. No new catalyst evident; lower priority.

To run a fresh dive on any of these, ask Meridian in the chat.