A fast-moving midday market wrap centered on Nvidia, SoFi, big tech, and AI infrastructure. The speaker argues that Meta’s expanded Nvidia relationship, Anthropic’s large cloud spend, and continued demand for Nvidia chips all reinforce a multi-year AI capex cycle, while also using SoFi’s 10-K to argue the lender’s credit quality is improving sharply. He also touches on prediction markets, the Fed minutes, Robinhood, Microsoft, eToro, Nebius, Shopify, and a broad risk-on tape.
Watch on YouTube ›Get the market thesis, key claims, assets, contradictions, and follow-up questions from any financial video — then unlock a version personalized to your portfolio, watchlist, and favorite speakers.
The speaker’s core thesis is that the AI infrastructure buildout is still early and still compounding, with Nvidia at the center of it. He says Meta has strengthened its relationship with Nvidia across networking, CPUs, Blackwell, and Vera Rubin, and he treats that as evidence that the market’s concern about Meta abandoning Nvidia for custom silicon or Google TPUs may be overstated. He ties that to Anthropic’s expected $80–90 billion of cloud spend through 2029 with Amazon, Google, and Microsoft, arguing that large customers are continuing to pay for cloud and Nvidia-powered infrastructure because it is the scalable, lower-total-cost option. He extends that same logic to a broader bull case for multiple AI-linked names: Nvidia, Micron, Nebius, CoreWeave, Amazon, Google, Microsoft, Meta, and even service providers like ServiceNow and Salesforce. …
Near term, this is a risk-on tape with Nvidia, cloud, and select software/fintech names benefiting from fresh headlines and earnings momentum. The main tactical risk is that crowded AI longs and high-beta growth names can reverse quickly if rates or sentiment turn.
Over the next several weeks and months, the base case is that AI capex and cloud spending stay supportive unless customers start diversifying away from Nvidia or the broader spend cycle slows. SoFi’s setup improves if the next data prints keep confirming better credit quality and operating leverage.
Structurally, the speaker believes AI infrastructure is a multi-year capital cycle centered on Nvidia and the hyperscalers, while prediction markets and compounding software/financial platforms may also gain secular importance. His long-run framework is stock-specific compounding rather than country or factor rotation.
Meta's strengthened deal with Nvidia disproves the overblown concern that Meta is moving to custom silicon or Google TPUs.
The speaker argues that Meta deepening its relationship with Nvidia across products shows fears about Meta shifting away from Nvidia were exaggerated.
Nvidia will trade above $250 per share this year.
Speaker references a personal bet with Steven Fiorillo that Nvidia will reach $250+ in the current year (February onward).
Prediction markets will become extremely large, eventually spawning prediction market funds and senior trader roles at major firms.
Speaker notes Susquehanna is hiring for senior prediction market traders building real-time models for event-driven outcomes, and argues this is becoming a very real market.
What do you think about prediction markets, especially with the CFTC defending them?
The speaker argues prediction markets are not going away and says the CFTC’s move reinforces that view. They think election markets were very useful, but are skeptical of some niche markets like five-minute crypto price contracts as genuine information tools.
What do you think about Robinhood's new venture fund?
They think it is fine, but they do not plan to participate. Their main concern is that these bundled funds reduce concentration and force investors into names they may not want, which can dilute returns on the companies they actually like.
Will you pay attention to Walmart's earnings today?
The speaker says Walmart’s earnings are not actually happening that day for them because they are on a different schedule. They then pivot to mention several other after-close earnings reports they find more interesting.
Unlock the full claims, asset map, scores, related transcripts, follow-up questions, and AI chat — shaped around your portfolio, watchlist, favorite speakers, and risks.
