IPOs and SPACs are Back, Mag 7 Showdown, Zuck on Tilt, Apple's Fumble, GENIUS Act passes Senate

Over the medium term (the next 5 years from June 2025), GPUs will remain the central compute architecture for training and running leading AI models; alternative architectures will emerge but will not displace GPUs as the primary platform.

“Look, I think to me, number one, I still think Nvidia right. I don't see the GPU kind of getting displaced. I see additional architectures kind of coming on board right. And growing the market. But, um, at the end of the day, all roads still lead to the GPU for all of these models.” View on YouTube

The video AI model referred to as "Vo 3" (likely Google's Veo 3) will cause such disruption that Hollywood, as an industry in its current form, will be effectively "done" within approximately one year from this June 2025 recording (by mid-2026).

“like Vo three, which we haven't really spoke about, is going to destroy Hollywood like in the next year, like Hollywood is done, I think.” View on YouTube

In the near term (within a few years from June 2025), China will develop advanced semiconductor manufacturing processes that create a significant competitive threat to Nvidia’s current dominance in AI chips.

“So I do think that there's going to be an emergent competitive threat coming out of China to Nvidia. And just like we were knocked over by deepfake, I think we will be knocked over by some semiconductor manufacturing processes, um, coming out of China in the near term.” View on YouTube

If Apple were to acquire OpenAI for around $500 billion, Apple’s stock price would rise on the day the acquisition is announced (the market reaction would be net positive).

“But I would love to see them be a little bit more aggressive. I mean, you guys remember when Steve Jobs bought Fingerworks, right? It was this tiny acquisition they made, this little trackpad that you could use your fingers on. No one figured out why they did this and then in turn into multi-touch and scrolling. Right. So I think it's it's going to be fascinating to see what they do.” View on YouTube

CoreWeave, Circle, Chime, and similar newly public high‑growth tech companies will emerge as the next generation of compounders, with many of them able to grow revenues at roughly 25% per year over the next 5–10 years.

“over the next 5 to 10 years, what are the companies that can compound and maybe 25% per year over that time frame? And I think companies like Core Weave and Circle and Chime, by the way, and others are going to kind of fill that gap.” View on YouTube

Over the coming years, essentially all major enterprise and business software systems globally will be rebuilt end‑to‑end using new AI‑enabled development toolchains, replacing the current generation of vertical SaaS and custom software.

“I think, Jason, if you look at the entirety of the software that runs the world, we're going to rebuild it soup to nuts. All of that” View on YouTube

Over roughly the next decade, within the S&P 493 (S&P 500 ex‑‘Mag 7’), there will be very large performance dispersion between companies that aggressively adopt AI to rebuild their business software and workflows and those that lag, creating some of the largest relative-return opportunities for public‑equity investors in decades.

“do you think that we enter an era where there is a similar dispersion, as we're talking about seeing in the Mag seven with the S&P 493, where there are going to be probably the biggest money making opportunities for investors that we've seen in decades between those that do adopt and do rebuild using AI and those that don't or are lagging.” View on YouTube

Chamath agrees that in the coming decade the S&P 493 will see very large dispersion in equity returns between AI adopters (which rebuild software and workflows with AI) and laggards, producing unusually large money‑making opportunities for stock pickers.

“100%.” View on YouTube

Traditional enterprise SaaS, sold as ever‑proliferating vertical tools with per‑seat or escalating license models, will structurally underperform going forward: growth rates will materially decelerate and return on equity will fail to meet prior expectations as customers increasingly reject ‘yet another tool’ in favor of AI‑driven, cheaper custom software.

“I think that the jig is totally up for software.” View on YouTube

Over the long term, pure consumption‑based pricing models for data platforms like Snowflake (where customers pay variably for large and growing data storage/compute) will prove unsustainable: many customers will migrate to lower‑cost alternatives (e.g., Postgres/Supabase and similar) and Snowflake‑style models will underperform or be forced to change.

“in this world, nobody's going to pay consumption because you're like, how do you expect me to, you know, hold and store and pay for terabytes and terabytes, potentially a day of data? It's not sustainable.” View on YouTube

In the next five years, public equity markets—starting with the Mag 7 and extending to the broader S&P 500—will exhibit unusually high dispersion between winning and losing stocks driven by AI adoption, making it an exceptionally attractive period for active stock pickers.

“for me as a stock picker, right. I think over the next five years, I couldn't think of a more interesting time where we're actually going to see dispersion between winners and losers.” View on YouTube

Many traditional ‘IT services’ and similar rolled‑up service businesses will face a lack of terminal buyers within roughly the next decade as AI agents become capable enough that much of their value proposition is automated away, depressing exit values for private‑equity roll‑ups in those sectors.

“I think the problem is that even if you take some of these kind of May industries and roll them all up, you ultimately have to find a buyer who wants to own that business after you...the fear that I have is that there is no terminal buyer for many of these companies.” View on YouTube

Over the next 1–2 years, the US IPO market will remain open for high‑quality growth tech companies (e.g., Figma and peers), with multiple ‘fantastic’ assets successfully going public and being well‑received by investors.

“I think we're going to see fantastic assets coming out. And I think the market is saying we're open for business...So bring on the new cohort.” View on YouTube

In the coming years of the AI transition, a strategy of shorting the broad S&P 500 while going long a small number of emerging ‘category killer’ AI beneficiaries will be unusually attractive and likely to outperform passive S&P 500 exposure.

“I think it's the first time you could probably argue that you could go short the S&P and pick a couple of winners...you could start to see category killers emerge out of the S&P.” View on YouTube

By roughly five years from now, Microsoft’s global employee count will be higher than it is today (around its ~250,000 peak), despite AI‑driven productivity gains.

“I suspect Microsoft's employee base grows.” View on YouTube

Within about five years, Microsoft’s total employee headcount will be lower than its recent peak of roughly 250,000, as revenue growth slows and AI and competitive pressures reduce the need for staff.

“I think shrink.” View on YouTube

Over the next five years, Microsoft’s overall business (revenue and scale) will be larger than today, and its total employee count will also increase, as growth in businesses like Azure more than offsets any AI‑related efficiency layoffs.

“I'm in Thomas camp, where I actually think the Microsoft business will be bigger if anything on on kind of usher alone. And that at the end of the day. Uh, we'll just need more people to support it.” View on YouTube

Over the next several years (on the order of five years), Microsoft will achieve very strong revenue growth while keeping its total employee headcount roughly flat around ~250,000 (±10%), leading to significantly higher revenue per employee.

“I'm predicting incredible growth and the same number of employees.” View on YouTube

An equal‑weighted index consisting only of AWS (if separable), Microsoft Azure, and Google Cloud Platform held over the next five years would deliver such strong returns that an investor ‘wouldn’t need to own anything else’—i.e., it would outperform the broad market and most diversified portfolios.

“If I could somehow automatically create an index of all three of those businesses right over the next five years...you wouldn't need to own anything else.” View on YouTube

Friedberg concurs that over the next five years, a portfolio effectively owning just the three major public‑cloud businesses (AWS, Azure, GCP) would outperform owning the rest of the market.

“100% own anything.” View on YouTube

Within a few weeks of this June 2025 discussion, the US House of Representatives will pass the Genius Act (the stablecoin bill), after which President Trump will sign it into law during his current term.

“And I would expect the House will act in the next few weeks on this. And then the president will have a bill he can sign.” View on YouTube

Following passage of the Genius Act, major US banks will launch their own US‑dollar stablecoins under the new regulatory framework.

“in the wake of this genius act, the stablecoin bill that the banks have now talked about getting into stablecoins, they're going to issue one.” View on YouTube

Under the Genius Act, Tether and other currently offshore USD stablecoin issuers will, within three years of the law’s enactment, relocate their stablecoin operations onshore to the United States and comply with US regulatory requirements.

“tether will under this act will have three years to come on shore. But the bottom line is they will have to operate in the United States.” View on YouTube

The referenced crypto-related bill that has passed the Senate will also pass the U.S. House of Representatives in the near term ("very quickly") following this episode's recording in June 2025.

“So hopefully this can pass the House very quickly.” View on YouTube

Later segments of this same All-In Podcast episode (beyond this transcript chunk) will include about 90 minutes of discussion on the Israeli conflict with Iran and Ukraine, featuring guests John Mearsheimer and Jeffrey Sachs in the second, third, and fourth hours of the show.

“We got two hours of classic all in. Uh, in part two of the show, we're going to do an hour and a half on the Israeli conflict with Iran. We've got 90 minutes. Ukraine. And we've got Ukraine, Ukraine, Ukraine. Mearsheimer and Jeffrey Sachs joining us in the second and the third and fourth hour of the All In podcast.” View on YouTube

The hosts (including speaker A and F) will meet in person on the upcoming Saturday night after this recording, in connection with the planned tequila launch event.

“Yes. We'll see you Saturday night. Absolutely.” View on YouTube