Miles Brundage, Executive Director of AVERI, says available evidence does not support claims of an outsized compute constraint at Anthropic and attributes outages to operational factors instead
Logan Graham of Anthropic confirms compute has not limited rollouts.
Obviously they’d be cooked without constantly making decibillion dollar compute deals
But they are constantly making decibillion dollar compute deals
Miles Brundage@Miles_BrundageI have not been convinced they have a huge compute problem yet. I mean everyone does to an extent in that there is scarcity etc. But I don’t think the Mythos rollout strategy was caused by this, and there are other explanations of things like outages such as sloppiness
#615Logan Graham@LOGANGRAHAM@Miles_Brundage Tried to be clear:
Logan Graham@logangrahamA lot of people have been wondering about Mythos, Glasswing, and the vulns we / our partners are fixing. Today, I’m excited for us to start sharing more. (For context, I lead Glasswing @AnthropicAI.) Two independent evaluations this week—from XBOW and the UK AISI—confirm what we've been seeing internally: Claude Mythos Preview is a step change in autonomous cybersecurity capabilities. We need to start preparing fast for a world of models with this level of capabilities. The UK AI Security Institute tested the model we shipped at the launch of Project Glasswing and found Mythos Preview is the first model to solve both of their end-to-end cyber ranges, including one (Cooling Tower) which no model had ever cleared. But attackers (and defenders) have sophistication & cost constraints – Mythos is also the only model that clears every one of their tasks estimated over 8 hours under their deliberately low 2.5M-token cap. XBOW tested it on their offensive security benchmarks, finding "token-for-token, unprecedented precision." It's the only model to succeed at subtle V8 sandbox work. Other Glasswing partners shared similar stories. In a few weeks of testing, Mythos Preview has helped them find many thousands of (estimated) high + critical severity vulnerabilities, sometimes double what they'd normally find in a year. I don't share this to boost Mythos. In fact, this is not about Mythos. It’s about preparing for the coming world of models being better, faster, cheaper, and more creative than some of the best human experts at dual use capabilities. Clearly, we need them supporting defenders as widely as can be done safely – and especially the least resourced ones. Within a year, Mythos will probably look quite dumb (relative to other new models). And others may release openly available or unguardrailed models of Mythos-level capabilities. We started Project Glasswing because capabilities like Mythos Preview's won't stay rare, or stay in careful hands. We are bringing it to defenders as fast as we responsibly can, while working to figure out, for example, the right safeguards and patching & disclosure processes. Also, to be clear, compute has never been a limiter in our rollout. Expect a fuller update on our Glasswing work in the coming days. XBOW report: https://xbow.com/blog/mythos-offensive-security-xbow-evaluation UK AISI report: https://www.aisi.gov.uk/blog/how-fast-is-autonomous-ai-cyber-capability-advancing
@Miles_Brundage one underevaluated thing is that it’s better to double your stock price and pay a 50% premium for compute than to buy up a lot of cheap compute early when your stock price is low. cost of capital is insane if your stock is undervalued and going up in the next few years.
Miles Brundage@Miles_BrundageObviously they’d be cooked without constantly making decibillion dollar compute deals But they are constantly making decibillion dollar compute deals
@Miles_Brundage secondly, a ton of cashflow (or cashflow equivalent) is still coming from investors. in this world, the point of revenue is to prove demand to investors, as much as it is to raise cash. so if you’re sold out due to demand - that’s mostly fine for raising the next round.
Ted Sanders@sandersted@Miles_Brundage one underevaluated thing is that it’s better to double your stock price and pay a 50% premium for compute than to buy up a lot of cheap compute early when your stock price is low. cost of capital is insane if your stock is undervalued and going up in the next few years.
More very popular AI coverage that's total dross. 400 upvotes on Hacker News but ignores:
• AI companies make nice gross margins (50%+) on average. • Cost to deliver the same AI performance is decreasing 10x per year. • You can always go down a model level and get performance that's ~90% as good for 25% the price. • Addressable market grew 6x from April '25 to April '26. That explosive growth makes it way easier to amortize fixed costs and achieve net profitability. • Yes the companies lose money on a few rare customers who figure out how to max out the $20 tier. But they're already mostly pushed up to the $50/$100/$200 tiers. And fundamentally they just don't matter much. Most on the $20 use very little.
Rob Wiblin@robertwiblinMore very popular AI coverage that's total dross. 400 upvotes on Hacker News but ignores: • AI companies make nice gross margins (50%+) on average. • Cost to deliver the same AI performance is decreasing 10x per year. • You can always go down a model level and get performance that's ~90% as good for 25% the price. • Addressable market grew 6x from April '25 to April '26. That explosive growth makes it way easier to amortize fixed costs and achieve net profitability. • Yes the companies lose money on a few rare customers who figure out how to max out the $20 tier. But they're already mostly pushed up to the $50/$100/$200 tiers. And fundamentally they just don't matter much. Most on the $20 use very little.