Briefing: Market Signals & High-Conviction Ideas

Purpose: Identify emerging market opportunities and high-conviction investment theses by monitoring macroeconomic indicators, sector momentum, and institutional positioning across equities — tracking options flow, earnings surprise magnitude, Fed policy shifts, and under-the-radar sector rotation signals.

Key Insights

• The rate environment has structurally shifted in a way equity markets are underpricing. The 30-year Treasury yield has broken above 5% for the first time since 2007, market pricing has flipped from two cuts to a 75% probability of a rate hike by year-end, and $9.7T in US debt refinances in 2026 at multi-decade high rates. The variant perception: most equity participants are treating this as an Iran-driven transient phenomenon, but bond markets are pricing something structural — Fed officials have explicitly flagged that AI-related memory chip shortages are adding 0.5% to core PCE independent of energy prices, making a peace deal insufficient to resolve the inflation problem. The actionable implication is a yield curve flattener posture and meaningful reduction in duration-sensitive growth stock exposure; the specific signal to monitor is the 2-year yield vs. Fed funds spread, which in the last 25 years has been associated with zero Fed cuts when in its current configuration.
• Bond Investors Are Panicking — And They May Be Right
• Markets Are Focused on Iran War. Peace Won't Fix the Inflation Problem.
• Stocks Halt Rally on Conflicting US-Iran Signals | The Close 5/27/2026

• America Just Repeated The Mistake That Crashed The Economy In 2008

• AWS has a structurally superior AI margin profile versus Google Cloud — and the market is missing it because of accounting. AWS margins improved 213 basis points quarter-over-quarter, driven by Anthropic's $30B ARR growing through Bedrock at a "$5.5B run-rate business today." The critical variant perception is the "EBTIT" framework: Google Cloud's reported margin improvement is an illusion because Gemini/DeepMind training costs are booked under "Alphabet-Level Activities," not within GCP — meaning Google's cloud margins are inflated by roughly the magnitude of its frontier model spend. Amazon's "token-as-a-service" model (the only cloud provider where this dominates rather than multi-year IaaS deals) means AWS captures a revenue share on each Claude inference, a structural advantage no competitor replicates. This is a cloud stock selection call — long AWS on relative margin durability, short or underweight GCP at equivalent reported margin multiples.

• Anthropic Growth and Bedrock Mix Drive AWS Margins Higher While Peers Lag

• The Lam Research options signal is the most actionable setup in the current batch. On a single day, 15,000+ OTM put contracts traded at 60x normal open interest — at a strike 8.34% below market with a $26.30 midpoint premium, generating a 9.07% yield over 86 days for put sellers. The SpotGamma framework clarifies why this is a bullish, not bearish, signal: when OTM put volume spikes to this magnitude, initiators are almost certainly put sellers setting a lower acquisition price while collecting income. The fundamental thesis supports it — FCF projected to surge to $11.6B on AI capex supercycle demand, implying 94% upside at the current 1.5% FCF yield. Key risk: if AI capex stalls or hyperscalers pull back on equipment spend, the semiconductor equipment cycle reverses sharply.

• Investors Love Lam Research Stock - Shorting OTM Puts in Huge, Unusual Volume

• Playing Against the House — The Hidden Power of Dealer Positioning | SpotGamma

• The AI infrastructure rotation has already moved from semiconductors into memory and is now entering fiber optics — institutional momentum flow confirms this sequence. Carson Block's internal momentum strategy (a basket of 20 stocks rebalanced monthly) has rotated through AI → Robinhood → memory → fiber optics in that order over the past 18 months, with fiber optics as the current leader. This is independently confirmed by fund flow data showing $20B+ into tech since the March low with all other sectors in net outflow, and by semiconductor ETFs dominating volume — suggesting the initial semiconductor trade is mature even as the market narrative remains fixated there. Fiber optic/optical networking names are the institutional momentum leaders per actual flow data, receiving far less mainstream coverage than semiconductors, and represent the cleaner current rotation signal.

• AI Risks Prompt Carson Block to Rethink India Plans
• AI Rally Gets a Reality Check From US Strikes on Iran | Insight with Haslinda Amin 5/28/2026

• Yahoo Finance Live: Daily Market Coverage - May 27, 2026 9AM-11AM (ET)

• The SLR exemption is an under-discussed policy escape valve that would constitute QE-through-banks if implemented. With US debt-to-GDP at 121% and the government unable to afford 1970s-style rate hikes, the most likely policy response is eliminating the supplementary leverage ratio for banks — the exact 2020 playbook. This would allow banks to purchase Treasuries without counting them against capital requirements, effectively suppressing yields and triggering a credit expansion boom. The timing signal to monitor: any Fed or Treasury language around bank capital rule relief. If implemented, the predicted outcome is a significant near-term asset price surge and economic boom, making the current risk-reduction impulse premature; if it fails to materialize, the rate burden on the fiscal position becomes increasingly acute.

• Inflation is EXACTLY Following the 70's - But They Can't Afford it This Time

Emerging Patterns

1. The AI infrastructure "winners vs. losers" bifurcation is becoming measurable in earnings data, not just narrative. Snowflake beat on Q1 revenue ($1.39B vs. $1.33B expected), raised its full-year product revenue guide to $5.84B (from $5.66B), and secured a $6B AWS deal for agentic AI chips — while Salesforce's Q2 revenue guidance came in at $11.27-11.35B against a $11.36B estimate, with remaining performance obligations missing and organic growth running at 6-8% against a 5-year double-digit guidance target. Jefferies' Brent Thill describes institutional clients as having been "incredibly negative" on Snowflake pre-quarter and now forced to reverse — a classic sentiment inflection following a guidance beat that exceeded the Q1 outperformance magnitude. The sector trade is explicit: overweight infrastructure (Snowflake, AWS, memory, fiber optics), underweight application software (Salesforce, Adobe), with cybersecurity firms that own system-of-record data as a protected middle ground. - Snowflake is a huge winner in AI, says Jefferies' Brent Thill - Salesforce Taking Longer Than Expected to Shift to AI, Analyst Luria Says - Salesforce CEO Marc Benioff goes one-on-one with Jim Cramer - Stocks Halt Rally on Conflicting US-Iran Signals | The Close 5/27/2026

2. The passive investing structural support mechanism for mega-cap equities has two simultaneous cracks forming. Carson Block identifies the market's supply constraint mechanism: passive inflows that never sell plus Mag 7 buybacks are the twin pillars keeping mega-cap valuations elevated. Both are now under stress simultaneously — hyperscaler AI capex is crowding out buybacks as cash flows are "starting to get drained," and AI-driven knowledge worker displacement (Block's "house view": 15% within 3-5 years) could eventually turn retirement fund inflows net negative, reversing the passive virtuous cycle. A third dynamic amplifies this: the SpaceX IPO and pending OpenAI/Anthropic IPOs will force index reallocation, redirecting capital away from current constituents (particularly Tesla) toward new entrants. This is a multi-year structural risk, not an immediate trading signal — but fiber optics as the current momentum rotation leader is the near-term expression of capital moving out of the initial AI infrastructure beneficiaries. - AI Risks Prompt Carson Block to Rethink India Plans - AI Rally Gets a Reality Check From US Strikes on Iran | Insight with Haslinda Amin 5/28/2026 - Yahoo Finance Live: Daily Market Coverage - May 27, 2026 9AM-11AM (ET)

3. Power infrastructure — not GPUs — is the real AI compute constraint, creating an under-owned "picks-and-shovels" layer. The Terawulf CEO states directly: "the real constraint now in the space is scalable power infrastructure" and "energy infrastructure is all" — companies that can secure hundreds of megawatts with guaranteed delivery are the true bottleneck. This is confirmed by the 800VDC datacenter transition thesis, which projects an $11B sidecar TAM peaking in 2028 and a $13B solid-state transformer TAM by 2030 — a technically-driven infrastructure shift largely un-owned by retail capital. The variant perception: market attention on GPU supply and semiconductor stocks misses that scalable power delivery is the harder, longer-lead-time constraint, and companies executing on energy infrastructure procurement have a defensible competitive moat the semiconductor names lack. - Terawulf CEO on demand in AI infrastructure - Inside the 800VDC Revolution – Part 1

Dissenting Views

The Salesforce bull/bear split is a genuine pass-or-act decision point, not a false controversy. The prevailing institutional view is bearish: Bank of America initiated with underweight citing "structurally lower growth" and transformation to "mature cash generator," executive departures (Denise Dresser to OpenAI) are flagged as a negative signal, and Jefferies recommends underweighting all application software. The dissenting bull case — made directly by CEO Benioff and supported by Cramer — is that Agent Force at $1B+ revenue with record million-dollar transactions, a $27B buyback ($46B full-year revenue guided), and a 12x earnings multiple represent a significant insider conviction signal against market skepticism. The resolution question is whether 6-8% organic Q2 growth is a transition trough before AI acceleration in H2 (Benioff's view) or a structural deceleration that the market "fully understands" at 10x cash flow (BofA's view). The specific variant perception to evaluate: does the CEO's aggressive buyback at these prices constitute informed insider accumulation or capital returned in lieu of investment because AI transition is slower than stated? - Salesforce CEO Marc Benioff goes one-on-one with Jim Cramer - Snowflake is a huge winner in AI, says Jefferies' Brent Thill - Salesforce Taking Longer Than Expected to Shift to AI, Analyst Luria Says

The SpaceX IPO creates a structured asymmetry between the pre-IPO sector momentum trade and the IPO itself. The mainstream narrative frames SpaceX as transformative and the IPO as a "historic" catalyst for space sector re-rating. The high-conviction dissent — from the Bond Investors Are Panicking source — is that 94x trailing revenue with $4.3B Q1 net losses and $10B annualized XAI losses constitutes a "completely insane" valuation, with a 70-80% peak-to-trough decline predicted within 24-36 months. Separately, governance concerns from Nell Mow at Value Edge Advisers are described as "the biggest I've ever seen" — Musk borrowing from the company, stock grants for unmet Mars colony targets, and index fast-tracking that perverts price discovery. The actionable separation: the sector momentum trade on liquid proxies (Rocket Lab as closest public proxy, AST SpaceMobile with volume 180% above average) is a legitimate catalyst play that works regardless of SpaceX's fundamental merit, while the IPO itself at proposed valuation is a specific pass with conviction. - Bond Investors Are Panicking — And They May Be Right - Squawk Pod: SpaceX governance & an AI film at the Tribeca Film Festival - 05/27/26 | Audio Only - 9 Space Stocks to Watch as SpaceX's Blockbuster IPO Approaches

Read & Act

What to Read

• Anthropic Growth and Bedrock Mix Drive AWS Margins Higher While Peers Lag — The "EBTIT" framework for Google Cloud is the most original piece of financial analysis in this batch and cannot be adequately summarized; you need to understand the specific deal structure (Anthropic/Bedrock revenue share) and the accounting treatment of GCP training costs to use this as a durable cloud stock selection tool. The 213bp margin inflection combined with the token-as-a-service business model distinction is the kind of variant perception that sustains a trade thesis across quarters.

• Playing Against the House — The Hidden Power of Dealer Positioning | SpotGamma — This provides the explanatory framework for why the Lam Research put signal is bullish rather than bearish, and more broadly why Salesforce dropped despite beating earnings. The put wall/hedge wall/gamma strike mechanics explain a wide class of "sell the news" outcomes in crowded names — directly applicable before any upcoming earnings catalyst in holdings like Nvidia, Micron, or Marvell where positioning is already extended.

• Bond Investors Are Panicking — And They May Be Right — This simultaneously delivers the most complete rate environment analysis, the sharpest SpaceX fundamental critique (with S-1 data showing Q1 net loss of $4.3B on $4.7B revenue), and the causal chain from yield increases through growth stock devaluation. The 62% of fund managers expecting 6% 30-year yield and Ferguson's Law on debt service vs. defense spending are arguments worth engaging directly rather than through summary.

• Inflation is EXACTLY Following the 70's - But They Can't Afford it This Time — The SLR exemption thesis is a specific, time-bounded policy catalyst with major market implications if implemented — it requires the full treatment to understand the mechanism (banks as QE proxy for the Fed), the 2020 precedent, and what language to monitor as a timing signal. This is the scenario most worth stress-testing against current portfolio positioning.

What to Do

1. Check OTM put/call volume ratios on your key semiconductor and AI infrastructure holdings before their next earnings catalyst. When OTM put volume hits 30x+ normal open interest, the SpotGamma framework indicates institutional put-sellers setting lower acquisition prices — a bullish signal, not distribution. Run this screen on LRCX, MRVL, and MU heading into their next reporting dates; if you see 60x+ spikes as appeared in Lam Research this week, that is an institutional conviction signal warranting a closer look at the put-selling entry structure rather than outright equity purchase.

2. Evaluate your cloud exposure specifically on the AWS vs. GCP margin distinction — not on aggregate hyperscaler exposure. The EBTIT analysis creates a specific relative value opportunity: AWS at structurally higher AI margins than GCP at equivalent reported margins. If your cloud exposure is expressed through a basket or equal-weight approach, this is worth disaggregating. The specific question to answer: what portion of your GCP valuation thesis rests on margin improvement that may be accounting-driven rather than operational?

3. Map your current AI infrastructure holdings against the momentum rotation sequence: semiconductors → memory → fiber optics. Carson Block's proprietary flow data shows fiber optics as the current institutional leader, succeeding the memory trade that succeeded the initial AI trade. If your AI infrastructure exposure is concentrated in first-wave semiconductor names that have already 50%+ outperformed (as the Philadelphia Semiconductor Index has year-to-date), evaluate whether trimming into fiber optic/optical networking names reflects where institutional capital is actually flowing rather than where mainstream narrative remains focused.