당신이 지켜보고 있다고 생각하는 AI 붐은 사실 중요한 그 붐이 아닙니다.

THE DAILY H.E.A.T. Torching Wall Street’s Obsolete Playbook. Delivered straight to your inbox 5 days a week. Subscribe Home Posts The AI Boom You Think You’re Watching Isn’t the One That Matters The AI Boom You Think You’re Watching Isn’t the One That Matters Mar 26, 2026 I’ve been a trader and investor for 44 years. I left Wall Street long ago—-once I understood that their obsolete advice is designed to profit them, not you. Today, my firm manages around $4 billion in ETFs, and I don’t answer to anybody. I tell the truth because trying to fool investors doesn’t help them, or me. In Daily H.E.A.T. , I show you how to Hedge against disaster, find your Edge, exploit Asymmetric opportunities, and ride major Themes before Wall Street catches on. Table of Contents The Final Frontier H.E.A.T. The AI Boom You Think You’re Watching Isn’t the On … News vs. Noise: What’s Moving Markets Today ETF News In Case You Missed It The Final Frontier We feel most thematic ETFs force you to choose between growth and income. We seek to provide both. SPCI targets concentrated economic exposure to the space industry while using a systematic put credit spread options strategy to aim for consistent cash flow. Our actively managed structure seeks to capture the sector's momentum while aiming to provide weekly distributions to shareholders. See how SPCI works. Distributor: Foreside Fund Services | Investing involves risk including possible loss of principle. H.E.A.T. The AI Boom You Think You’re Watching Isn’t the One That Matters Wall Street is still debating chip cycles and GPU benchmarks. That’s the wrong argument. The real story is the shift from AI that answers questions to AI that does the work — and it’s rewriting the profit pool across software, security, cloud, and infrastructure. The Narrative Phase One of AI is over. Ask a question, get an answer. Summarize a document. Draft an email. That was Phase One — impressive, mostly passive, and already priced by the market. Phase Two is something different. It’s delegation. Agents don’t autocomplete — they execute. Build the deck. Pull the data. Write the code. Fix the bugs. Route the approvals. Submit the report. And do it again tomorrow without forgetting what you like. The kicker: you don’t need to code anymore. You “code” in plain English. That’s why the agent race is turning into a land grab. OpenAI says Codex has crossed 2 million weekly active users, with Reuters reporting it crossed 1 million developers in a single month. Anthropic’s Claude Code is scaling fast enough that it’s been cited as generating billions in annualized revenue. Whether you love these companies or hate them, the message is the same: agents are turning “work” into “compute.” Citi projects that AWS AI revenue will account for 72% of incremental growth by 2027. BofA just reinstated coverage on CoreWeave, Microsoft, and Oracle specifically on the AIaaS thesis. And Wells Fargo flagged Claude Code’s impact on IT budgets as “The Next Debate in Software.” The consensus is no longer “who has AI?” It’s “who survives agents?” “Agents are turning ‘work’ into ‘compute.’ The market is shifting from chatbots to autonomous workflows — and it’s creating a brutal bifurcation between moated platforms and commoditized point solutions.” — Synthesis of BofA, Citi, WFC, JPM research consensus Here’s the paradox that’s starting to hit markets: agents increase AI usage while potentially shrinking traditional software usage. If a company can do the same output with fewer people, it may need fewer seats of seat-based SaaS, fewer junior analysts, fewer “workflow clicks” across ten different apps. That’s why the “SaaSpocalypse” narrative got traction. But here’s where the crowd gets it wrong: software doesn’t disappear. It consolidates. The market opportunity expands — but the rent shifts toward platforms that own the workflow, own the data, and own the control plane. In other words: the agent economy doesn’t kill enterprise software. It kills undifferentiated software — the stuff that’s basically a nice UI on top of a set of obvious features. 2M+ Weekly active Codex users (OpenAI) OpenAI / Reuters 72% AWS AI share of incremental revenue by 2027 Citi Research Phase 2 Agentic AI — from answering to executing Market consensus 3 Qtrs BRZE consecutive accel. organic growth Company filings / JPM The Proof Point THE PROOF POINT: BRZE — Why Braze Is the Clearest Signal in the Room Today Three consecutive quarters of accelerating organic growth. JPMorgan has an AI-focused expert call scheduled specifically to examine Braze. And the thesis is simple: Braze is a system-of-record for customer engagement data — permissioned, workflow-embedded, and irreplaceable by a generic agent. Agents need data to act. Braze owns the data. That’s not a moat. That’s a drawbridge. What the numbers say: → 3 consecutive quarters of accelerating organic revenue growth → AI features driving upsell — not cannibalizing ARR → JPM expert call signals institutional focus is building → Positioned at the intersection of agents + enterprise data + compliance The verdict: BRZE is the rare software name where agents make the product more valuable, not less. That bifurcation is the entire trade. Winners & Losers ✓ Winners ✗ Pressure Points Tier 1 — The Toll Collectors Risk isn't company failure. It's pricing model failure — fixed seats, thin margins, no unique data, and a generic agent that does the same job for a fraction of the cost. Nvidia NVDA Agents are inference-hungry. More agents = more GPUs, more memory bandwidth, more power. The cleanest picks-and-shovels play in the stack. Zoom ZM Seat-based, feature-thin, easily approximated. Agents reduce the need for scheduled coordination — the core use case is under pressure. Broadcom AVGO Networking silicon and custom AI accelerators — AVGO sits at the intersection of hyperscaler capex and agent infrastructure. DocuSign DOCU Single-surface workflow app. Agents can route, sign, and track approvals without a dedicated platform. Repricing risk is real. Arista Networks ANET AI clusters need extreme networking. Inference at scale drives bandwidth demand that legacy switching can't handle. Box BOX Cloud storage layer with no unique data moat. Agents don't need a dedicated storage UI — they call APIs. Consolidation target. Micron Technology MU Memory is the silent bottleneck of agentic AI. HBM demand is structural, not cyclical. NVDA can't run fast without MU. Asana ASAN Project management without deep workflow entrenchment. Agents can manage task routing natively. Seat math breaks down fast. Tier 2 — The Control Plane CrowdStrike CRWD Agents need permissions. Permissions create attack surface. Security spend doesn't fall when AI rises — it accelerates. Labor arbitrage models face structural pressure: outsourced back-office, basic research, entry-level analytics, low-end content. The 'headcount always rises' assumption is broken. Palo Alto Networks PANW Zero trust becomes mandatory when agents are touching sensitive enterprise systems 24/7. PANW owns the perimeter. Okta OKTA Identity and access management is the control plane for what agents are allowed to do. Every agent deployment is an OKTA opportunity. Tier 3 — The Consolidators Microsoft MSFT Copilot + Azure + enterprise surface area. MSFT is the closest thing to an agent operating system for the Fortune 500. BofA reinstated on this exact thesis. Watch out for 'AI app' companies selling $200/mo magic while burning $600/mo of compute. Margin is dumb even when the model is smart. The unit economics will surface in 2025 earnings. ServiceNow NOW Owns enterprise workflow. Owns enterprise data. Owns enterprise procurement cycles. Agents run ON ServiceNow, not instead of it. Braze BRZE 3 consecutive quarters of accelerating organic growth as AI enhances — not destroys — its value. The clearest proof point of the 'moated platform' thesis. JPM expert call confirms institutional focus is building. The bifurcation trade in one ticker. What Would Break This Thesis What would break this thesis ∧ Agent adoption stalls: enterprises hit compliance, liability, or hallucination walls that slow deployment beyond 2026 — buying time for seat-based SaaS. ∧ Compute costs collapse faster than expected: if inference gets 10x cheaper rapidly, the 'inference subsidy' phase extends and masks which business models are truly broken. ∧ Regulation: EU AI Act enforcement or U.S. executive action constrains autonomous agent deployment in regulated industries — finance, healthcare, legal. ∧ The macro breaks first: if a recession hits before agents reach scale, the entire capex build slows and 'agentic infrastructure' becomes a 2027 story instead of a 2025 one. Key Takeaways 01 We are at the Phase 1 → Phase 2 inflection right now. Chatbots were impressive. Agents are operational. That shift moves spending from ‘AI features’ to ‘AI infrastructure’ — and the two have very different profit pools. 02 The toll collectors win regardless of which agent platform dominates. Compute (NVDA, AVGO), memory (MU), networking (ANET), and security (CRWD, PANW) get paid every time an agent runs. They don’t need to pick a winner. 03 The control plane is not optional. Agents need permissions. Permissions create attack surface. Every enterprise agent deployment is a cybersecurity and identity management event. Security spend accelerates, it doesn’t soften. 04 BRZE is the clearest bifurcation trade in software right now. Three consecutive quarters of accelerating growth as AI enhances its value. It owns the data agents need to act. That’s not a moat — that’s a drawbridge. JPM’s focused expert call is institutional confirmation the thesis is building. 05 Agents don’t kill software. They kill weak software economics. The question for every software position in your portfolio is simple: does this company own the workflow, the data, or the control plane? If the answer is no — it’s a feature waiting to be absorbed. News vs. Noise: What’s Moving Markets Today I gave my thoughts on the markets to the NYSE yesterday at the open…… Bottom line, until this war is over you cannot expect the market to do a lot on the upside. Peace talks are better than boots on the ground, but they don’t mean the war is over….. Iran suspects Trump's peace talk push is another trick "Trump has a hand open for a deal and the other is a fist waiting to punch you in the f***ing face." www.axios.com/2026/03/25/iran-war-trump-peace-talks-tehran-suspicious So far this morning oil and rates are back up, remember, that’s the tell. Gameplan remains the same, find the areas that were strong before the war and buy them on dips. Mix it in with names that maybe got unfairly beat up and are making moves back up. Both sides need this to end, hopefully that means something at some point. ETF News MEMY Holdings Update: We sold Palantir PLTR and Astera Labs ALAB and bought Seagate Technology STX and Almonty Industries ALM. All 5% positions. For a full list of MEMY holdings, visit: https://incomeblastetfs.com/etf/memy Distributor: Foreside Fund Services, LLC A Stock I’m Watching Today’s stock is JFrog (FROG)…. The stock had an undercut and rally at the 10 and 20 day moving averages the other day, would like to see it break back through the 50 and 200 to confirm a new uptrend. Not really a fan of software at the moment, but FROG is an exception…. The Quiet Backbone of the AI Software Supply Chain While the market chases flashy AI names, JFrog is doing the unglamorous but mission-critical work that every serious software organization depends on: securing and distributing the code that runs modern applications. JFrog's Artifactory platform is the de facto standard for binary repository management — essentially the warehouse and shipping dock for software artifacts — and as enterprises accelerate AI-driven development cycles, the velocity and volume of code being built, tested, and deployed is exploding. More code shipped faster means more demand for JFrog's platform, which sits at the intersection of DevSecOps and MLOps. The company has been methodically layering in security capabilities (runtime security, secrets detection, software composition analysis) that make it increasingly sticky and hard to rip out. With a land-and-expand model, strong net retention, and a growing presence in cloud hyperscaler marketplaces, FROG is the kind of name that doesn't need a headline catalyst — it just keeps compounding relevance. It's not the loudest stock in the room, but in a world where AI makes software supply chains both more complex and more vulnerable, JFrog is quietly becoming essential infrastructure. In Case You Missed It I had the pleasure of talking to Dividend Degenerates on why I like put spreads better than covered calls for income….. The H.E.A.T. (Hedge, Edge, Asymmetry and Theme) Formula is designed to empower investors to spot opportunities, think independently, make smarter (often contrarian) moves, and build real wealth. The views and opinions expressed herein are those of the Chief Executive Officer and Portfolio Manager for Tuttle Capital Management (TCM) and are subject to change without notice. The data and information provided is derived from sources deemed to be reliable but we cannot guarantee its accuracy. Investing in securities is subject to risk including the possible loss of principal. Trade notifications are for informational purposes only. TCM offers fully transparent ETFs and provides trade information for all actively managed ETFs. TCM's statements are not an endorsement of any company or a recommendation to buy, sell or hold any security. Trade notification files are not provided until full trade execution at the end of a trading day. The time stamp of the email is the time of file upload and not necessarily the exact time of the trades. TCM is not a commodity trading advisor and content provided regarding commodity interests is for informational purposes only and should not be construed as a recommendation. Investment recommendations for any securities or product may be made only after a comprehensive suitability review of the investor’s financial situation.© 2026 Tuttle Capital Management, LLC (TCM). TCM is a SEC-Registered Investment Adviser. All rights reserved. Keep Reading Mar 20, 2026 • 13 min read Bezos Just Filed for a 51,600‐Satellite “AI Data Center” Network... Here’s the Real Trade Mar 18, 2026 • 12 min read Washington Just Declared War on China’s Shipyards: 3 Stocks That Could Benefit Mar 17, 2026 • 8 min read Why I Look at What Analysts Can't Cover View more 7