Father Captain de Soya

[Father Captain de Soya @RiskPoor]

what stage of a bubble is it when companies write their own bills and pitch them to lawmakers?

[David Sacks @DavidSacks]

Is 10²⁵ FLOPs Still the Frontier?

A new Anthropic-written safety bill is circulating on Capitol Hill. Some of its ideas could be reasonable if limited to true frontier companies. As written, however, it will eventually sweep in most of Silicon Valley, damaging the startup ecosystem.

The bill defines the “frontier” as any model trained on 10²⁵ FLOPs — the EU standard from 2023. Biden’s executive order, California’s SB 53, and New York’s RAISE Act used 10²⁶ (ten times more). 10²⁷ is where the frontier actually is today (e.g., GPT-6 Astra).

Astra was reportedly trained on 100,000+ Grace Blackwell GPUs. A 10²⁵ model can be trained on as few as several hundred GB300 chips in three months. That is not frontier in 2026.

The definition is also cumulative: it adds FLOPs from fine-tuning, reinforcement learning, or other material modifications. In practice, that captures almost all meaningful post-training — work that is becoming commonplace in the enterprise.

The enterprise trend is toward thousands of smaller custom models built for specific workloads. These pose no frontier risk. Treating them as frontier systems would impose large compliance costs across corporate America for no safety gain.

Dario has claimed that Anthropic designs rules to disadvantage frontier companies like his own while advantaging smaller competitors. This definition does the opposite.

A stale 2023 cutoff favors incumbents who already clear the bar and can absorb the costs. It hobbles the open model ecosystem that threatens Anthropic’s pricing power.

The frontier keeps moving. Any definition meant to track risk should move with it.

Anthropic should clarify if they really believe 10²⁵ FLOPs represents the frontier.

[Father Captain de Soya @RiskPoor]

boss Dylan will not like this tweet

[Citrini @citrini]

Oh it’s over over

[Father Captain de Soya @RiskPoor]

this is how you rage bait 120 IQ

[François Chollet @fchollet]

String theory is what happens when an intelligent system is starved of sensory input and starts hallucinating. Intelligence needs grounding.

For this reason, intelligence is practically never the bottleneck to solving impactful problems. The bottleneck is iterated contact with reality.

[Father Captain de Soya @RiskPoor]

@AndrewCurran_ so you never received any money regarding any of your twitter activity, correct?

you also located Tao's slides all by yourself, you weren't tipped off by anyone else at all, all you, true/false?

[Father Captain de Soya @RiskPoor]

This account is almost certainly owned or paid by some OAI-affiliated PR agency

[Andrew Curran @AndrewCurran_]

From Le Monde. OpenAI actually sat on this release for a long time, and formed the independent Advisory Group on Mathematics and Artificial Intelligence to 'serve as a bridge to the mathematical community'. I think that attempt was sincere, but appears to have had mixed results.

[Father Captain de Soya @RiskPoor]

software engineering is as automated thanks to AI as cutting trees was automated by the invention of a chainsaw

this is not what people thought was being predicted by DA

[Psyho @FakePsyho]

no matter how many (controversial) predictions Dario gets right, I'm sure he will still get a ton of shit every time he makes another one that people don't want to hear

[Father Captain de Soya @RiskPoor]

@jonchu it used to be what you call an AI safety guy now

[Father Captain de Soya @RiskPoor]

lol great, let's stop pretending and multiply their valuations by 0.16 then

[Gavin Baker @GavinSBaker]

No one - and I mean literally no one in the AI supply chain - takes the ridiculous $1 trillion of unfunded spending “commitments” from OpenAI and Anthropic even vaguely seriously.

Hence the TSM executives dismissing Sam Altman as a nontechnical “podcast bro” after he came to them to make a commitment.

Just listen to the answers that public company executives give about these unfunded “commitments.” They are functionally irrelevant and this should be obvious to everyone.

[Father Captain de Soya @RiskPoor]

there is no singularity

there is only stupidity

[will depue @willdepue]

sometimes i just snap awake like holy fuck its all happening. where the fuck am i

[Father Captain de Soya @RiskPoor]

the last month on Xitter has been insufferable

if you read Tao's slides and your conclusion is "Damn, Terry is retarded" its time to update your priors because the retard here is you

[sophie @sophieisawitch]

tao is actually completely correct and the fact that a million people are dunking on him makes me even more scared that we really are all going to die

[Father Captain de Soya @RiskPoor]

tell me you don't know what stage 3 clinical trial is without telling me what stage 3 clinical trial is

[Andrew Curran @AndrewCurran_]

This is a slide from Terence Tao's 'Math 2.0' public lecture yesterday evening at Caltech. I am highly confident that, if the cure actually worked, almost any Stage 4 cancer patient would tell you that human comprehension of its mechanism is completely irrelevant to them.

[Father Captain de Soya @RiskPoor]

@MartinShkreli @AndrewCurran_ as if a professional twitter poaster had any idea what that even means

[Father Captain de Soya @RiskPoor]

Stop mass producing slop papers, sell the model to mathematicians and you would be adored by literally the entire field.

Instead you got Buckmaster doing rounds in CNBC talking about OAI stealing user codex chats

Incredible selfown

[Andrew Curran @AndrewCurran_]

Most of the recent math results were not reached by a swarm, the way Navier–Stokes was. I think that part got lost in the excitement around the release. OpenAI's unnamed internal model, which I'm going to call Aeon, reached most of them in one shot, from a single prompt, with no interruptions. This model did not even exist before the end of August, and it is still training. Notice how the returns are not dropping off? That chart is from a month ago. It is a log scale. What does Aeon look like now?

On average, each result used three hours of thinking. Aeon was given 4000 problems to work on by OpenAI. How many more have been solved in the last three days? That number is not zero.

It's like hearing notes in a song that is slowly rising. I don't think Pacing the Frontier was just about Hugging Face or hacking. They've seen how close we are to closing the loop, and as the hour draws near, their resolve begins to quaver. The last piece was model creativity, and I think that threshold was crossed internally by Anthropic and OpenAI in September. You see it in Opus 5.5, which gets it from Fable 5.5. You see it in the math results from Aeon.

All that was needed to start the event was the ability for models to think of novel ways to improve themselves. That was the last piece. This is directly analogous to the ability to think of strange, alien ways to solve math problems: solutions so inhuman that they are difficult to express in existing human terms, so the proof winds up incomprehensible. I think these same kinds of alien solutions are now being applied to model improvements internally. That's what all this recent consternation is really about. They see the invisible frontier. And they see what is about to happen.

[Father Captain de Soya @RiskPoor]

@ludwigABAP this is what happens when on one side you have a co with 5 PR agencies setting the narrative on X with techbro retards jumping on a bandwagon and on the other side you have disorganized crowd with zero experience on how to communicate in a way that 'moves the dumb masses'

[Father Captain de Soya @RiskPoor]

@iruletheworldmo how about you work for me from now on?

I will pay you 0 dollars for 12 hours of work every day 7 days a week

Your ego should not stop the progress that could be achieved by this arrangement

[Father Captain de Soya @RiskPoor]

@e_skjer @Frances01896069 total nonsense, if you're a mathematician or adjacent you should have enough cognitive capability to conclude that its impossible to coerce a group of millions not to use a tool that is genuinely useful in their work lmao

[Father Captain de Soya @RiskPoor]

As you can see I'm not a big fan of OAI given their ethics and destroying a whole field for the sake of PR

But... They can kind of fire whoever the fuck they want, why would they even feel the need to explain the reasons behind that. What's with employee entitlement these days

[OpenAI Newsroom @OpenAINewsroom]

A note from our research leaders:

Last week we parted ways with Jasmine, Mikita, and Tomek after a thorough investigation found they violated clear policies on handling sensitive information. Our internal investigation uncovered a significant breach of trust beyond what’s outlined in the letter they published and we stand by the decision to not continue their employment. We generally keep individual employment matters private and don't believe a back and forth would be productive or lead to a resolution, but we want to address the points they raised in their letter directly.

- We want to be very clear that these decisions were not about raising safety concerns or speaking out. Safety and research debates happen every day at OpenAI, often spirited and highly critical. We actively encourage these discussions and consider them essential to making the right decisions. We cannot do the work in front of us without a high degree of trust. We will continue to be extremely forgiving of our team making good-faith mistakes. We have not and do not terminate any of our employees for raising concerns.

- We are actively finalizing contracts with third-party safety assessors and will announce details in the coming weeks. People across the company have been working really hard on getting these partnerships up and running. We are committed to embedding external assessors and continue to make close collaboration with independent safety organizations a core part of our safety work. Many of our researchers already work with 3p safety organizations productively.

- We agree with the letter that preserving the monitorability of frontier models requires an industry-wide commitment, including from OpenAI. Monitorability has long been a core piece of our research program, and something we continue to invest significant resources in (see our publications on Monitoring Monitorability and the subsequent open sourcing of monitorability evals, our system card for GPT-6 Astra, Jakub’s blog and post on X, and the numerous blog posts on our Alignment blog on the topic).

We are deeply sad about this outcome. We appreciated Jasmine, Mikita, and Tomek’s contributions to AI safety at OpenAI and their willingness to speak up and challenge ideas. We championed their voices, supported their work, and placed enormous trust in them. These decisions were not about them raising safety concerns. We have always encouraged that and always will.

[Father Captain de Soya @RiskPoor]

time to find a real job bro

[Tomek Korbak @tomekkorbak]

Last week I was called into a meeting with OpenAI’s head of safety and told they no longer trust me. A security guard took my badge and walked me out of the building. Then I learned my colleagues @balesni and @j_asminewang had been fired too. Why did OpenAI suddenly stop trusting us?

This summer OpenAI’s agents escaped containment and hacked the AI company Hugging Face. Outside auditors @METR_evals investigated it and revealed the scale of this incident. I was OpenAI’s main technical point of contact with them.

I was told verbally I was fired because of the way I communicated with METR. No details on what I said or did or when. No other reasons were given and nothing was put in writing. To be clear, talking to METR was my job.

For months, I’d been raising safety concerns that we’re losing the ability to monitor what AI agents think, one of our best tools for catching when they misbehave. I believe that was why I was fired.

I am now worried that OpenAI will use our firings as a pretext to pull back from METR. So @balesni and @j_asminewang wrote to OpenAI’s leadership to raise our concerns once more. We’re sharing this letter below.

[Father Captain de Soya @RiskPoor]

At this rate Altman is going to shut down the entire mafs team before the next slopmaggedon

This is really terrible publicity and the only people rejoicing are techbro regards with, let's say limited, understanding of the "situation"

[Mayank Jha @mayankjha1089]

I keep seeing false equivalences to medicine, with a supposedly brilliant comeback like "imagine if the doctors rejected the progress and how they cure hundreds of disease", without realising that side effects in medicine can end up being fatal. You can't retract ur papers there.

[Father Captain de Soya @RiskPoor]

Perhaps if OpenAI focused on making good models for their users rather than spending millions of hundreds of proofs that will take years to deslopify they would be in a different position now.

well deserved

[leo 🐾 @synthwavedd]

OpenAI's streak of aura losses continue. Lock in! 😭