A fast-paced two-person market discussion centered on SoFi, Micron, Nvidia, and CoreWeave/Nebius. The speaker argues SoFi’s bull case is not broken but the “tech platform” narrative has weakened, while Micron and the AI infrastructure complex look increasingly attractive because consensus may be underestimating the duration and size of the capex/compute cycle.
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.
This transcript is a conversational market debate between Steve and Tanner that starts with personal banter and quickly turns into portfolio and thesis discussions. The main focus is whether to sell SoFi to buy Micron. Steve argues he is not selling SoFi, but he is increasingly frustrated that SoFi is being valued like a tech platform when, in his view, it is functioning more like a bank. He says his original thesis—SoFi as the “AWS of fintech”—has been damaged because recent results weakened revenue diversification and the platform/business-services story, even though he still believes the company can continue growing and hit guidance. A large section is devoted to comparing SoFi with traditional financial institutions such as Citigroup and JPMorgan Chase. …
Near term, the action is in AI-linked names and Nvidia’s upcoming earnings, while SoFi looks more like a thesis digestion period than an immediate sell. Micron and CoreWeave are the more active “new money” ideas in this conversation, but both remain volatile and estimate-sensitive.
Over the next few months, the base case is that AI capex and compute demand keep supporting semis and neoclouds if hyperscaler spending stays strong. SoFi can still compound, but unless it reasserts a platform narrative, the market may continue valuing it as a fast-growing financial institution rather than a tech hybrid.
Structurally, this conversation argues that AI infrastructure is becoming a multi-year capex regime spanning chips, memory, networking, and cloud capacity. The lasting implication is that companies with durable pricing power and visible demand in that stack may deserve much higher strategic attention than traditional valuation heuristics would suggest.
Steve is not selling SoFi to buy Micron; he is using cash to buy Micron instead.
He explicitly corrects the title framing and says he is not selling SoFi.
SoFi’s original “AWS of fintech” thesis has been damaged by the latest quarter.
Steve says his thesis is broken even though the bull thesis isn’t fully broken.
SoFi is increasingly just a bank in the market’s eyes, and that will limit the multiple unless a tech component re-emerges.
Steve repeatedly says SoFi is now being valued like a bank and needs a tech component to earn a premium.
Should we sell SoFi to buy Micron?
Steve answers no: he is not selling SoFi and is instead using cash to buy Micron.
Why did you think I was going to sell SoFi?
Steve says the original SoFi platform thesis was hurt by the quarter, especially around revenue diversification, Galileo, and the banking-as-a-service story, but he still is not selling.
What made you get into CoreWeave?
Steve says valuation is the main reason, along with strong AI demand, Nvidia support, scale, and perceived cheaper relative valuation versus Nebius.
Unlock the full claims, asset map, scores, related transcripts, follow-up questions, and AI chat — shaped around your portfolio, watchlist, favorite speakers, and risks.
