We are going to see a lot of vertically focused AI native companies accelerate. Routers, open-source models and specialized post-training enabled by companies like @FireworksAI_HQ have all made dramatic advances and the combination of the three is driving accelerating growth. Companies like @wearelegora can now use their data to post-train an open-source model and then combine it with frontier models behind a router to get the same or better outcomes at lower costs than the frontier alone. This dramatically improves the business model for all these companies. @cognition seeing similar trends.

A series of fairly wild thoughts from Dwarkesh. I thought I was bullish but even I am not assuming that the price to rent compute continues to go up! I will say that I am in SV this week and Dwarkesh is capturing the zeitgeist. Net, net public markets would probably be trading differently if they saw the OpenAI, Anthropic, Grok/Cursor and Open Source numbers over the last 6 weeks. And really good for anyone who has installed compute coming off contract and/or is bringing on compute that is not already contracted. And credit slowing down capacity adds - if it happens - only exacerbates all of this.

Market is overreacting to hyperscale credit spreads widening from my perspective. TL;DR Spot pricing for renting GPU compute materially above contracted rates implies hyperscalers are underearning while operating cash flow acceleration is an underestimated source of funds for AI capex. The fact that spot prices for GPU rentals are at least 2x higher than contracted rates is the missing piece from the discussion about hyperscaler credit, which is the only fundamental factor behind this selloff. Multiple private companies are planning on spending at least 2x more per GPU for compute as contracts roll-off and some have spoken about this publicly.   As contracts roll-off, hyperscale growth rates are going to continue to accelerate as their installed bases of compute reprice higher. Hyperscale operating cash flow growth using a mix of estimates and actuals is modeled to accelerate from 31% in the first quarter of 2026 to 50% in the second quarter. This acceleration should continue for the rest of the year and this is not in estimates which incorrectly model a deceleration in the third quarter from my perspective.   Some math. Consensus estimates are probably for 25-35 gigawatts added by hyperscale and neoclouds in CY28 (using a range as standing up datacenters is hard and a lot of the neos plus labs are still private).  At 60b per gigawatt, that is 1.5 to 2.2 trillion in capex. Consensus estimates for hyperscale/neo operating cash flow is 1.3 to 1.4 trillion. I think this gets revised up materially as contracts reprice and growth accelerates so the 100b to 700b that would hypothetically need to be plugged by debt goes away. And their credit profiles materially improve. Not to mention the said 100b to 700b would be less than 1 turn of incremental leverage on consensus EBITDA estimates. And obviously the Nvidia and Broadcom “credit wrappers” help improve creditworthiness as well given their FCF profiles.   OpenAI, Cursor/Grok and the various Open Source inference clouds have accelerated materially over the last two months per public data and Anthropic continues to grow insanely fast while likely generating FCF. This - along with the fact that spot prices for GPU rentals are so far ahead of contract - are the missing pieces from the BofA chart on hyperscale FCF vs. semiconductor FCF.   Hyperscalers are underearning and anyone who signed a contract for GPU compute in 2024 and 2025 is overearning. Operating cash flow will be enough to fund capex but as contracts reprice and cloud growth continues to accelerate then spreads likely come in as well.  
Would also note that CDS markets are easy to manipulate - was a huge feature of the GFC - short the stock and then buy the CDS. So I would not put attach much signal to CDS. 
Net, net I’m not that concerned about the widening spreads in hyperscale credit. The real risk is that bringing power online and energizing all these GPUs is really hard but we are getting better at this every day.

An outcome where AI is dominated by a few dominant frontier models is dangerous for America. Super negative for national security. The Hugging Face incident and Anthropic/DoD dustup are object evidence of the risks. Happy to see this.

“38x slowdown once shuffle data exceeds available DRAM and spills to SSD.” From a joint Micron/Meta white paper. The DRAM bottleneck almost certainly lasts the longest. OP from @jukan05

New Pareto Frontier: Grok 4.5, SWE-1.7 and Opus 5. Intelligence per $ will be the only metric that matters over time. K3 probably joins the frontier once available on the inference clouds.

Kimi K3 may be an important inflection point for AI. Potentially negative for Anthropic and OpenAI while being net positive for essentially every other company in the world. I mean that very literally. Although the real “Sputnik moment” would be an open-source frontier model that was also token efficient unlike Kimi K3 which is 50-70% more expensive to run than GPT 5.6 per Artificial Analysis. Rationale:   A world where there are only 2-3 dominant frontier labs with 90% inference margins is net negative for every other layer while being awesome for those 2-3 labs. Those labs would become monopsonies for power, data centers, semiconductors and hyperscalers and would obviously vertically integrate over time into all those layers while also completely subsuming the application/software layers.    Anything that lowers margins and increases competition at the model layer is good for every other AI layer: power, semiconductors, hyperscalers, neoclouds and yes even software.   This is why Jensen is so supportive of open-source. An open-source model requires the *exact* same amount of compute to run as a closed frontier model of similar size and architecture. Kimi K3 is roughly the same price as GPT 5.6 Terra on a per token basis, which actually suggests that it is less computationally efficient as I am sure that GPT 5.6 is priced to a higher margin than K3. And given that K3 is a token wastrel, i.e. token inefficient, it is significantly more expensive per task than GPT 5.6 and Grok 4.5, which are much more token efficient. Cost per token and token efficiency (i.e. intelligence density per token) are the drivers of intelligence per unit of cost. The winning AI companies will be those that offer the most intelligence per $ over time.   Lower margin % at the model layer = more margin $ at every part of the infrastructure layer and is a godsend for software. This can happen either through open-source models like K3 at the frontier *or* having a vertically integrated model company like Meta, SpaceX or Google at the frontier. Both outcomes result in a lower margin % at the model layer as vertically integrated model companies don’t really care where the margin $ come from. This is why it was so painful for OpenAI and Anthropic when Google was right there with them from a model competitiveness perspective and why Grok 4.5 and Muse 1.1 were just as important as Kimi K3. 
The reason Kimi K3 is only *potentially* negative for Anthropic and OpenAI is 1) the @ericvishria point that the Claude and ChatGPT products and harnesses may be more important than their models today and 2) the hypothesis that they have much more advanced model checkpoints internally that are already being used for RSI. In the latter scenario, reaching RSI even a few months ahead of other labs might be enough to cement a permanent lead. Time will tell on both points. And likely fairly quickly. Caveat would be that since Kimi K3 is not token efficient and thereby actually more expensive than ChatGPT 5.6, we may need to see a more token efficient open-source model at the frontier or see Grok 5/Composer 4/Muse 2 at multiple points on the Pareto frontier for this potential risk to Anthropic and OpenAI to play out. And I am sure they will both vertically integrate as quickly as possible while continuing the product/harness strength they have shown over the last 8 months.

The mega bull case for AI infrastructure would be *if* market share shifted away from certain frontier labs with 90%+ inference margins toward cheaper models, whether open-source or closed. It would increase the ROI on AI spend for end customers by increasing intelligence per dollar, which would drive incremental token demand. Margin dollars would effectively get redistributed from the frontier labs to AI infrastructure providers. The infra winners would be those with the lowest per token cost and the winners at the model layer would be those with the highest token efficiency. There are many reasons Jensen is so focused on open source, but this is likely the most important one as I think he is probably less worried about a monopsony these days. Lower margin % at the model layer = more margin $ at the infra layer all else equal. With SpaceX and Meta being vertically integrated and possessing the #3# and #4# models respectively it is more possible than ever. Note that Grok 4.5 is ahead of Fable for some useful tasks at a much lower cost, so ranking them #3# is conservative. This is not happening yet. Cheap, mostly open source tokens are likely the majority of volume today but the majority of economic value is still accruing to the most intelligent models. Might change though. We will see.

Great day @nvidia And today I learned that Jensen’s favorite food is Chicken Fried Steak, which he had soon after arriving to Kentucky as a 9 year old boarding student who cleaned the school bathrooms. 🇺🇸 🇺🇸 🇺🇸 🇺🇸 🇺🇸

I hope that Atreides and Valor are able to secure the franchise rights to @McDonalds on both the Moon and Mars. And Valor has actual expertise running franchised restaurants! Who can help make this happen? I will put a deposit down today unless SpaceX wants the rights.

@AntonioGracias @reindsummit @McDonalds More photos like this please. Also kinda funny that you went from correcting Bloomberg’s reporting about Valor’s SpaceX ownership in your last post to fried apple pie in the next. The American Dream in action! And who doesn’t love both McDonalds and SpaceX? As 🇺🇸 as it gets!

Kinda wild how wrong most energy experts were about the Straits of Hormuz. What were the root causes of the error? China had more reserves than anticipated, US flexed up exports faster and more ships were actually transiting the straits via the Omani route? Genuinely curious.

Such a fun day with so many old friends. Always a little sad to leave Starbase. Per aspera much of the time so important to celebrate sometimes. So grateful to everyone at SpaceX for helping to make the future awesome. Excited for the next Starship launch. Ad Astra!