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 $5 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
H.E.A.T.
~8% Global DRAM revenue share (2025) | 300k Current wafers/month; 600k planned | $2.9B Reported Tencent server DRAM deal | 29.5B¥ IPO raise target (Shanghai listing) | ~30% China smartphone LPDDR share |
CXMT was supposed to be a consumer-memory nuisance. It is becoming a server-memory supplier.
Reuters reported the other day that Tencent has signed a multiyear DRAM supply agreement with CXMT worth more than 20 billion yuan — roughly $2.9 billion. The agreement reportedly covers server DRAM supply over several years. CXMT and Tencent have not confirmed the details, and the scope of HBM inclusion remains unclear. That uncertainty matters. But the signal is hard to miss.
A Chinese DRAM supplier that Wall Street still treated as a low-end commodity player is now reportedly tied to one of China's largest cloud and AI companies.
This is not a story about CXMT beating SK hynix in HBM3E. It is a story about China building a domestic memory floor under its own compute stack — while Samsung, SK hynix, and Micron are busy chasing the HBM profit pool.
The Trojan chip is no longer just inside a laptop. It is moving toward the server rack.
THE TIMELINE November 2025: Caixin Global reports CXMT supplying DDR5 to Lenovo. March 2026: CXMT chips confirmed inside shipping Lenovo ThinkBook 2026 units. February 2026: HP and Dell reportedly begin qualifying CXMT DRAM. June 29, 2026: Reuters reports CXMT signs $2.9B multiyear server DRAM supply deal with Tencent. Each step was dismissed as incremental. Together, they are a structural shift. |
PART I: THE GEOMETRY OF THE THREAT
Let's be precise about what CXMT is and is not, because the investment implications live entirely in that gap.
CXMT is not the world's best DRAM company. Samsung's 1c-node process and SK hynix's HBM3E dominance are not threatened by a Chinese foundry running DUV multi-patterning at a node the Big 3 surpassed in 2018.
But here is the thing about commodity markets that Wall Street perpetually misunderstands: you do not need to be the best to reshape the price structure. You just need to be good enough, at scale, at the moment when your competitors are busy elsewhere.
And the Big 3 are unambiguously busy elsewhere. Samsung, SK hynix, and Micron are all reallocating wafer capacity toward HBM at an accelerating rate. HBM dies are substantially larger than commodity DDR5 dies. The same wafer yields fewer units. The arithmetic is unavoidable: as HBM allocation goes up, commodity DRAM supply tightens.
Into that tightening supply gap walks a company with 300,000 wafers per month of capacity today, plans to double that to 600,000, and a Tencent server contract now anchoring its customer list.
"CXMT does not have to beat the Big Three at the frontier. It only has to become good enough at domestic scale while the Big Three chase HBM." |
This is the geometry of the threat. Not a frontal assault on Samsung's technology leadership. A flanking maneuver into the gap the Big 3 are voluntarily vacating — and Tencent just confirmed the flank is open.
PART II: PRODUCT BY PRODUCT — WHERE THE GAPS ARE REAL AND WHERE THEY ARE CLOSING
LPDDR5X (Mobile): Already a Market Force. CXMT holds roughly 30% of China's smartphone LPDDR market. Xiaomi, Oppo, Vivo, and Honor are shipping products with CXMT memory at scale. In February 2026, Qualcomm CEO Cristiano Amon confirmed on the Q1 FY2026 earnings call that Qualcomm is in a qualified state with all memory suppliers, including CXMT. The technical pathway for Chinese OEMs to standardize on CXMT memory is open.
DDR5 (PC): Shipping, Not Sampling. CXMT's G4-node DDR5 is now inside Lenovo products. The die size is approximately 67 square millimeters — roughly 40% larger than Samsung's equivalent per TechInsights analysis, implying a meaningful cost disadvantage. But the product ships and lands in customer hands. The gap between a lab sample and a Lenovo purchase order is not a minor engineering milestone. It is the entire game.
Server DRAM (RDIMM/MRDIMM): No Longer Theoretical. The Tencent deal upgrades this category from possible beachhead to reported long-term supply agreement. Reuters lists Tencent, Alibaba Cloud, ByteDance, Lenovo, and Xiaomi among major customers in CXMT's IPO prospectus. Global hyperscaler certification cycles are long and unforgiving — CXMT is not breaking into AWS racks. But Alibaba Cloud and Tencent Cloud operate under a different incentive structure. Supply chain independence is a strategic imperative, not a procurement preference. That market is now open.
HBM: No Frontier Credibility — But That Is the Wrong Question. CXMT is not a credible frontier HBM supplier to Nvidia-class global AI platforms today. Reuters reported in 2024 that CXMT had developed sample HBM chips with Tongfu, and that CXMT aimed to begin HBM3 mass production in 2026, with a dedicated HBM back-end packaging fab under construction in Shanghai. The relevant question is not whether CXMT can compete with SK hynix HBM3E. It is whether CXMT can supply good enough memory for Huawei's Ascend accelerators, Cambricon's chips, and second-tier Chinese inference deployments that cannot rely on US-controlled supply chains. On that question, the answer is increasingly yes.
PART III: 3D DRAM — THE TAIL RISK NOBODY IS PRICING
The conventional DRAM competitive order rests on a single foundation: EUV-enabled process node shrinks. Samsung, SK hynix, and Micron have invested tens of billions in EUV infrastructure. CXMT cannot access EUV due to export controls. Therefore CXMT is permanently behind. That is the consensus view.
3D DRAM does not invalidate that consensus view today. But it puts a question mark next to it.
3D DRAM — stacking memory cells vertically rather than shrinking them laterally — changes the competitive calculus because the primary driver of density in a vertical architecture is not lateral lithography. It is etch depth, deposition uniformity, bonding precision, and yield management across stacks. These are areas where DUV-dependent manufacturers are not structurally disadvantaged in the same way they are in planar scaling.
The NAND flash parallel is instructive. In 2013, 2D NAND hit a physical wall. The planar leaders had spent decades optimizing lithography. When 3D NAND emerged, it did not extend the old leaders' advantage — it reset the competitive order. Companies that moved fastest on vertical architecture gained the most ground, regardless of prior planar position.
A senior SK hynix Americas executive and a current ASML employee, speaking separately at a Silicon Valley Korean semiconductor meetup in March 2026, both flagged China's 3D DRAM efforts as deserving more attention than Western analysis affords them. We treat this as directional, not confirmed. Seoul National University Professor Hwang Cheol-seong has separately assessed that CXMT could produce 11nm-class DRAM without EUV.
CREDIBILITY NOTE The SK hynix and ASML executive comments are directional observations from named-role sources at an industry event — not official disclosures. Professor Hwang's assessment is a published academic analysis. No public technical data on CXMT's 3D DRAM capability has been disclosed. We are treating 3D DRAM as an asymmetric tail risk, not a base-case thesis. See Credibility Firewall below. |
The right framing: 3D DRAM is not the base case. It is the tail risk that makes the CXMT story more dangerous than a normal low-end capacity addition. If density starts moving vertically instead of laterally, export controls built around EUV scarcity become less complete. That does not mean CXMT catches SK hynix tomorrow. It means the "permanently behind" assumption deserves a lower confidence interval.
PART IV: THE EQUIPMENT ECOSYSTEM — NEAR-TERM TAILWIND, MEDIUM-TERM POLICY RISK
The most important investment insight from the CXMT story is not about CXMT itself. It is about what CXMT's aggressive capacity buildout does to equipment demand — and who captures that demand over time.
ASML — Narrative Watchlist, Not Fundamental Pressure Yet. CXMT cannot buy EUV tools. ASML's moat spans leading-edge logic, foundry, an enormous installed base, and future high-NA. None of that is threatened by CXMT near-term. The risk is subtler: if credible 3D DRAM scaling emerges outside EUV's reach, the "EUV is the only density path" narrative premium gets questioned at the margin. That is a multiple conversation, not an earnings conversation — but multiples move first.
Lam Research and Applied Materials — Near-Term Tailwind, Watch Policy. CXMT's $6-7 billion in 2023-2024 capex, and a new fab two to three times the size of its existing footprint underway, represent incremental demand for etch and deposition equipment not currently restricted by BIS controls. Near-term order book is a tailwind. The medium-term concern: if CXMT scales 3D DRAM using DUV-compatible processes, it validates a roadmap that concentrates purchasing power in Chinese fabs — purchasing power US policy may eventually constrain more aggressively.
Tokyo Electron — The Cleanest Near-Term Read. As a Japanese supplier exposed to both legacy DRAM process tools and 3D DRAM-relevant deposition and etch systems, TEL is not subject to US export restrictions in the same way. It is the most direct beneficiary of CXMT capex among the named equipment names — and the most transparent signal for tracking CXMT fab buildout pace.
The Domestic Equipment Layer — The Gap in Most Western Analysis. BIS's December 2024 controls expanded restrictions on advanced semiconductor manufacturing equipment including etch, deposition, lithography, metrology, inspection, cleaning, and HBM-related items. That makes the long-term equipment story less simply "TEL/Lam/AMAT get orders" and more "CXMT pulls forward both foreign-tool demand and Chinese substitution demand simultaneously." The direct public-market read-through includes China's domestic tool ecosystem: Naura, AMEC, and ACM Research. CXMT's capex is a foreign-tool tailwind in the near term and a domestic-substitution accelerator over the medium term. For US-listed investors seeking exposure to China semi equipment without A-shares, ACMR is the most liquid proxy.
Western OEM Risk — Reportedly Qualifying, Not Confirmed Adopting. Reuters reported in early 2026 that HP had reportedly started qualifying CXMT DRAM and might use it for non-US markets if supply tightness persisted; Dell was reportedly qualifying because of rising price fears. Acer and Asus were reportedly exploring local sourcing through Chinese production partners. None of these are confirmed adoption decisions. They are procurement options becoming real — and that is sufficient to change the Big 3's pricing behavior.
WINNERS & LOSERS
COMPANY / SECTOR | VERDICT | WHY IT MATTERS | RISK |
SK hynix | NEAR-TERM WINNER | HBM3E dominance unchallenged. CXMT threat concentrated in commodity space hynix is already deprioritizing. Any CXMT domestic server share is largely irrelevant to hynix's AI memory thesis. | 3D DRAM acceleration by CXMT would threaten longer-term node leadership narrative; any HBM supply shock from Chinese domestic sourcing |
Tokyo Electron | WINNER | Direct beneficiary of CXMT DUV buildout with limited US export restriction exposure. Most transparent read on CXMT fab buildout pace. 3D DRAM ramp adds incremental etch/deposition demand. | Policy escalation restricting Japanese suppliers; 3D DRAM validation timelines uncertain |
Lam Research | NEAR-TERM WINNER / MEDIUM-TERM WATCH | CXMT capex means near-term etch tool order flow not currently restricted. Each new CXMT fab adds incremental backlog. | US export control expansion to cover more etch/deposition categories reverses the tailwind immediately |
ACMR (ACM Research) | DOMESTIC BENEFICIARY | US-listed liquid proxy for China semi equipment exposure. Directly in line for CXMT's domestic substitution capex as export controls tighten on foreign tools. | US policy targeting US-listed China-linked companies; customer concentration risk in CXMT build cycle |
ASML | NARRATIVE WATCHLIST | EUV monopoly intact. Near-term fundamentals unchanged. Not a direct CXMT loser. Risk is multiple compression if 3D DRAM scaling outside EUV creates credible density alternative. | If 3D DRAM proves competitive with planar EUV scaling, justification for premium valuation gets complicated — a multiple question, not an earnings question |
Micron Technology | CONTESTED WINNER / CHINA PRESSURE | CXMT pressures commodity DRAM and China exposure, but HBM and broader DRAM shortage can still support Micron earnings. US memory sovereignty narrative is a tailwind. Risk is not immediate collapse; it is multiple compression if investors stop treating commodity DRAM tightness as structurally protected. | Commodity DRAM oversupply scenario; China market share erosion accelerates; pricing floor cracks before HBM mix improves |
PC OEMs (HP, Dell, Acer, Asus) | STRATEGIC OPTIONALITY | Reported CXMT qualification gives supplier negotiating leverage. BOM cost reduction meaningful in a thin-margin hardware business. Non-US market optionality is real. | Geopolitical risk: CXMT adoption triggers secondary sanctions exposure or customer perception issues in US government segments |
Applied Materials | NEAR-TERM NEUTRAL / WATCH | Similar near-term profile to Lam. Non-restricted tool categories benefit from CXMT buildout. Longer-term depends on 3D DRAM architecture choice and policy trajectory. | Policy risk on deposition tool exports; single-customer concentration lumpiness if CXMT dominates incremental DRAM capex |
PRESSURE POINTS
PRESSURE POINT | WHAT TO WATCH | TIME HORIZON |
Tencent Deal Scope | Reuters could not confirm whether HBM is included. A follow-on disclosure including HBM would materially reprice CXMT's AI-infrastructure relevance. | Q3 2026 |
CXMT IPO Filing | Prospectus will contain yield data, customer concentration, and capex guidance not currently public. This document moves stocks on the day it drops. | 2026–2027 |
HP/Dell Formal Qualification | Any confirmed Tier 1 global OEM qualification announcement is a commodity DRAM pricing event, not just a supply chain story. | Q3–Q4 2026 |
Alibaba / ByteDance Server Deals | If Tencent is first, others follow. Watch for additional Chinese cloud and internet companies formalizing CXMT supply agreements. | H2 2026–2027 |
3D DRAM Demonstration | Any public technical demonstration of competitive 3D DRAM by CXMT or a Chinese foundry reprices equipment stocks immediately. | 2027–2028 |
Export Control Expansion | US tightening of DUV multi-patterning restrictions or etch/deposition tool rules slows CXMT buildout materially. BIS December 2024 controls already expanded scope. | Ongoing — event-driven |
HBM Back-End Fab Completion | CXMT's dedicated HBM packaging facility in Shanghai determines whether it can credibly supply China's domestic AI hardware demand. | 2027 |
CREDIBILITY FIREWALL
COMPANY-DISCLOSED | MODEL-DERIVED | EDITORIAL VIEW |
CXMT holds ~30% of China's smartphone LPDDR market (CXMT disclosure, Caixin Global) | Global DRAM revenue share ~8% in 2025; Q1 revenue 50.8B yuan, up ~700% YoY; net profit 25B yuan (Reuters citing IPO filing) | Revenue is catching up to capacity faster than the market modeled — Q1 profitability is the structural inflection point |
CXMT DDR5 confirmed shipping in Lenovo ThinkBook 2026 series (Caixin Global, March 2026) | CXMT DDR5 die size ~67mm² vs. Samsung equivalent per TechInsights (~40% cost disadvantage implied) | Cost disadvantage is real but not permanent — in a supply-constrained market, it doesn't need to be eliminated to matter |
Reuters reports CXMT signed >20B yuan / ~$2.9B multiyear server DRAM supply deal with Tencent; CXMT and Tencent did not comment (Reuters, June 29, 2026) | Tencent, Alibaba Cloud, ByteDance, Lenovo, Xiaomi listed as major customers in CXMT IPO prospectus (Reuters) | Unconfirmed deal details and HBM scope are the key unknowns. We treat the server supply relationship as reported but not officially confirmed. |
CXMT IPO filing (Reuters) shows 29.5B yuan raise target; prior reports cited $42B valuation; ~2.3B yuan net loss H1 2025 now reversed per Q1 2026 profitability | $6–7B capex 2023–2024; new Shanghai fab reportedly 2–3x existing size; HBM back-end fab under construction (Reuters/IPO filing) | The company crossed to profitability in Q1 2026. IPO timing is no longer just a capital story — it is a credibility validation event. |
Qualcomm confirmed qualified status with CXMT on Q1 FY2026 earnings call (February 2026) | HP reportedly qualifying CXMT for non-US markets; Dell reportedly qualifying due to price fears (Reuters citing Nikkei Asia) — not confirmed adoption | OEM qualification discussions are a leading indicator. Confirmed adoption follows 12–18 months after technical qualification. |
SK hynix Americas head and ASML employee flagged 3D DRAM concerns at Silicon Valley Korean Semiconductor Meetup (March 2026 — directional, not official) | No public technical data on CXMT 3D DRAM capability disclosed; Professor Hwang Cheol-seong (Seoul National University) assessed CXMT could reach 11nm-class without EUV | Treating 3D DRAM as an asymmetric tail risk is the correct posture. It is not a base-case thesis and should not be presented as one. |
BEAR CASE: WHY THIS COULD STILL BE NOISE
BEAR CASE SPOTLIGHT The Tencent deal is unconfirmed and the companies did not comment — the reporting could be incomplete, delayed, or overstated. CXMT's die size disadvantage means it burns more wafer area per unit than the Big 3 — structurally worse economics until process parity arrives. Server qualification cycles are long and unforgiving, and the Tencent relationship may be exploratory rather than operational at scale. HBM is where AI infrastructure spending concentrates, and CXMT has no credible frontier HBM offering today. If global DRAM demand disappoints, CXMT's aggressive capacity doubling becomes a liability, not an asset — and the newly achieved profitability evaporates. Export controls, while imperfect, create genuine long-term headwinds that DUV multi-patterning alone cannot solve forever. |
FIVE THINGS TO DO WITH THIS INFORMATION
1. Reread the Tencent deal carefully when details emerge. Reuters could not confirm whether HBM is included. If a follow-on disclosure confirms HBM content in the Tencent supply agreement, this reprices CXMT's relevance to China's AI infrastructure stack — and puts pressure on the HBM-only thesis for SK hynix and Micron.
2. Watch CXMT's IPO prospectus as the most important single document in the memory investment thesis for 2026. Yield data, customer concentration, capex guidance, and product mix will all be disclosed publicly for the first time. This document will move equipment and memory stocks on filing day.
3. Monitor Tokyo Electron's order book guidance for the cleanest real-time read on CXMT fab buildout pace. TEL's non-restricted exposure to DRAM process tools makes its quarterly disclosures more transparent than Lam or AMAT given ongoing policy uncertainty around etch and deposition categories.
4. Add ACMR to the watchlist as the most liquid US-listed proxy for China semiconductor equipment demand. As BIS controls tighten on foreign tools, CXMT's capex increasingly flows toward domestic substitutes. ACMR is the clearest public-market expression of that dynamic without requiring A-share exposure.
5. Stop anchoring on CXMT's revenue share. The relevant number was always capacity share — low-to-mid teens globally — and the story has now moved past capacity to confirmed customer relationships. Tencent as a server DRAM customer is a different kind of proof point than Lenovo as a PC customer. Revenue follows customer relationships. Institutional coverage follows revenue.
The AI Buildout Has a Physical Layer
Many of today’s data centers are still using copper wiring. The same metal we’ve been using for a hundred years.
At the speeds AI demands with data moving between thousands of GPUs, billions of times a second, copper doesn’t just slow down.
It turns that data into heat. The more you push through it, the worse it gets. There’s no software for fix for that.
So what’s the answer?
Explore the Photonics Layer…..
Tuttle Capital Pure Play Photonics ETF (FOTO)
Distributor: Foreside Fund Services | Investing involves risk including possible loss of principle.
News vs. Noise: What’s Moving Markets Today
The news. The semiconductor trade is not dead. It is getting narrower. That is different. Micron just told you the memory shortage is real: $41.46 billion of quarterly revenue versus $35.85 billion expected, $22 billion of customer supply commitments, and management saying tight conditions can last beyond 2027. Apple reportedly wants permission to buy memory from blacklisted Chinese supplier CXMT (see above) because memory costs are hitting Mac and iPad pricing. Applied Materials hit a record high Monday after analyst target hikes. Goldman says AI infrastructure stocks are expected to drive nearly 60% of S&P 500 EPS growth this quarter, with Nvidia and Micron alone making up roughly 40%. That is not a broad market story. That is a bottleneck story. The parabolic phase paused. The scarcity phase did not. Sources: Reuters, MarketWatch, Investor’s Business Daily, Barron’s.
The noise. The market wants to price the Strait of Hormuz like the crisis is over. It is not over. It is paused. Barron’s reported oil settled higher Monday after the U.S. and Iran agreed to halt strikes and meet Tuesday. Benzinga, citing NYK’s Takaya Soga and the FT, said Hormuz traffic may stay below 50% of prewar levels for months because safe routes are “extremely limited.” That is the Red Sea lesson. A ceasefire is not a shipping lane. A memo is not demining. Insurance, mines, drones, Iran’s leverage over routing, and Israel/Lebanon all still matter. The other noisy headline is the yen at 162.41, the weakest since 1986. That is being framed as Japan’s problem. It is bigger than that. Japan already spent 11.7 trillion yen trying to defend the currency, and Reuters says the market is pricing a 61% chance of a Fed hike by September. If the carry trade unwinds, today’s calm VIX is not telling the truth. Sources: Barron’s, Benzinga, Reuters, Cboe.
The dumb advice. The dumb advice is treating AI earnings strength as a green light to buy the whole index. Goldman’s Ben Snider may be right that AI infrastructure will carry Q2 earnings. The problem is what that actually means. If two stocks are 40% of S&P 500 EPS growth and AI infrastructure is nearly 60%, that is concentration, not confirmation. Reuters put it better: AI is working for the providers, not the spenders. Hyperscalers have issued $60 billion of bonds in the last 12 months, Mag Seven names are down more than 10% by one measure, and investors are starting to question free cash flow and ROI. That is why the broad-market victory lap is dangerous. Own the toll collectors. Be careful with the capex funders. And don’t pretend a 30-year Treasury at 4.854% is a safety blanket if Warsh’s Fed is still inflation-dominant. Sources: Barron’s, Reuters, MarketWatch, MarketWatch 30-year.
Concrete takeaways.
Semis are paused, not broken. Buy strength in the bottlenecks, not every AI ticker.
Memory is still the cleanest AI shortage signal. MU, HBM, storage, equipment, and power matter.
Hormuz is not resolved. Treat the “stand down” as a trading headline, not a shipping solution.
Yen weakness is the sleeper risk. Intervention may pop the yen, but the carry trade is structural.
Long bonds are not a safe haven here. They are a macro bet on Warsh blinking.
The market is broadening, but earnings power is still concentrated. That is opportunity and risk.
Where Does the Money Go When AI Hits a Wall?
When capital chases a tech theme, it tends to pile into the most obvious
layer and miss the one underneath. AI spending is now bumping hard
against memory. Hyperscalers — the big cloud builders like Amazon,
Google, and Microsoft — have shifted memory from 8% of their build
budgets to an estimated 30% in a single cycle. That capital has to go
somewhere. If the constraint is memory, and the build can't move without
it, shouldn't an investor own the layer AI runs on?
View HBMX fund holdings →
Distributor: Foreside Fund Services | Investing involves risk including
possible loss of principal.
<Link = http://www.hbmxetf.com/>
ETF News
A Stock I’m Watching
KTOS may be a buy as a sympathy/second-derivative drone trade off AVAV because AVAV’s blowout quarter shows investors are willing to re-rate unmanned-defense names again: AeroVironment reported record Q4 revenue of $641.6M, fiscal-year revenue up 141%, $2.7B in bookings, and $1.2B of funded backlog, while calling out rising demand across drones, counter-UAS, space, and advanced defense technologies. (AeroVironment, Inc.) KTOS fits that same theme but has its own fundamental support: Q1 revenue grew 22.6%, Unmanned Systems organic revenue grew 30.9%, Valkyrie-related activity drove segment growth, and total backlog reached $2.01B, including $1.46B funded, giving the company meaningful revenue visibility. (Kratos Defense) The setup is that drone stocks have been washed out, AVAV’s earnings can pull attention back to the group, and KTOS offers exposure not just to drones but also Valkyrie CCA, jet engines for drones, missiles, loitering munitions, hypersonics, and broader defense recapitalization themes. (Kratos Defense) While valuation is not cheap, the combination of renewed sector momentum, accelerating unmanned-systems demand, backlog support, and rising defense-budget urgency makes KTOS a credible catch-up candidate if the AVAV move sparks a broader bid in drone and next-gen defense names.
In Case You Missed It
Great conversation on wide ranging topics with Kenny Polcari…
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.