One can instead require that “AI R&D cannot be conducted or assisted by AIs more capable than those trained 9 months ago”, with capability being operationalized by something like ECI and measured by embedded evaluators. This would allow restarting the GPT-7.5 RL from an earlier checkpoint without “resetting the clock”.
brendanhalstead
Karma: 1,099
i also heard this from what i believe is an independent source at OAI.
[Note that we recently published the code, web access is disabled right? it shouldn’t be in the pretraining yet, but will be soon...]
brendanhalstead’s Shortform
The code for the AI Futures Model is now published here: https://github.com/AI-Futures-Project/aifm-public .
i have the impression that they are.
Q2.5 2026 Timelines Update: Uplift and Revenue
How to pace the US frontier
The original post on the METR blog has an interactive version of this figure where you can vary,, and.
this link does not exist
The model for capabilities progress is described here https://ai-2040.com/supplements/takeoff-supplement . It seems like the base rate of crazy discontinuous architectural improvements that work at much-lower-than-frontier-levels-of-compute is low. I feel like I’d be more worried about them if there was an argument for why the underlying inputs to that discovery process would be much higher in Plan A, otherwise “people have been able to think about brain-like architectures for quite a while, but haven’t succeeded yet” seems like evidence that it’s quite hard. Maybe advances in actual neuroscience would make it easier?
Who are you imagining would “[produce] superintelligence long before?”
AI R&D making use of legal compute is highly regulated and totally transparent.
AI R&D making use of illegal compute could happen, but we think a secret AI project like this (1) would have a hard time staying undetected (2) would have a very hard time using the “top human expert” legally-deployed AIs for AI R&D undetected (due to inference monitoring) (3) would be limited to ~30x less compute than the leading labs had before the deal, even assuming high competence in setting up an illegal chip diversion program in advance of the deal. According to our current best-guess modeling, the chance that a covert project like this would reach TED-AI undetected before 2040 in this scenario is around 10%. (For comparison, we think the chance that the deal would dissolve or become significantly degraded by that point is more like 50%.)
For the math, the assumption that might not hold today is that the elasticity of research effort to compute is constant over a large range like 10x.
For the simulations in the AIFM, it changes the results slightly. In the image at the bottom, the top left graph shows the effect of a pause + reduction two years before Automated Coder. Looks like the median drops to about 5x.
AI 2040: Plan A
How much slower does takeoff go with 10× less compute?
3-4 GW at the end of 2026
Are you basing this on this quote?
There’s a few strings to pull on there. One is, what happens to depreciation of GPUs? I guess I didn’t answer your prior question, which is that I think Anthropic will be able to get to five gigawatts-ish, maybe a little bit more by the end of the year through themselves as well as their product being served through Bedrock, Vertex, or Foundry. I think they’ll be able to get to five or six gigawatts, which is way above their initial plans. OpenAI will be roughly the same, actually a little bit higher based on our numbers.
added a link
These results seem to support the hypothesis that 5.5 is a new pretrain, though what’s happening with 5.3 and 5.4 is a bit unclear.
[edit: code available here, credit belongs to @Daniel Paleka for the idea]
- 's comment on Model Size Scaling in 2023-2031 by (23 Jun 2026 1:28 UTC; 14 points)
- 's comment on Vladimir_Nesov’s Shortform by (2 May 2026 15:07 UTC; 9 points)
- 's comment on Vladimir_Nesov’s Shortform by (27 Apr 2026 20:56 UTC; 5 points)
- 's comment on Vladimir_Nesov’s Shortform by (30 May 2026 13:46 UTC; 5 points)
- 's comment on Vladimir_Nesov’s Shortform by (29 Apr 2026 18:30 UTC; 2 points)
Q1 2026 Timelines Update
If I were Anthropic, I’d be worried about distillation attacks via this route. It seems to make their CoT obfuscation moot.
But couldn’t the third party evaluators just ask “what are all your AI R&D benchmarks?” and make an ECI-like thing out of them? They can’t bench-min on the same benchmarks they’re using internally to assess progress, right?
“intentionally training the models to sandbag” also seems catchable by embedded evaluators or whistleblowers