X日报 · AI科技

Introducing a limited preview of GPT-5.6 Sol, our next generation frontier model, as well as GPT-5.6 Terra, a balanced model for efficient, everyday work, and GPT-5.6 Luna, a fast and affordable model for high-volume work.

https://t.co/OoM83SyISN

Since June 12, we’ve been working closely with the US government to restore access to Claude Mythos 5 and Fable 5. Today, the government notified us that Mythos 5, our strongest cybersecurity model, can be redeployed to a set of US organizations that operate and defend critical infrastructure.

We’re restoring access for these organizations quickly, and we’re continuing to work with the government to expand access to Mythos 5 and make Fable 5 available for general use again.

Good new first: Sol is a smart, efficient, and a significant step forward. It is the same price as GPT-5.5. Also launching in the GPT-5.6 family is Terra, with 5.5-level performance at half the price.

Bad news: at the request of the US government, it is launching today in limited preview instead of the open access launch we were planning on. We are working with the government to get to general availability as fast as we can.

I think it is quite reasonable to roll out models--especially as they reach significant new levels of capability--in this way. It fits with our long-held strategy of iterative deployment. But this isn't quite the process that we think is optimal.

Now we will with the government to attempt to get to a transparent, reliable process for early access, and to ensure that as long as our safeguards work as intended we can release widely. We want to be a reliable, dependable partner that works with all stakeholders, and we also want to live by our mission of benefiting all of humanity. I believe the government shares most of our goals, and that they are overall doing a good job in a very difficult situation.

We will work as quickly as we can to get this model in your hands and we hope you will love it.

We want to help all companies be secure, working with the USG and the security ecosystem.

*The full version of GPT-5.5-Cyber is here; state of the art performance on CyberGym.

*Patch The Planet and Codex Security will help solve security problems instead of just finding them. https://t.co/otyCFHJR4d

Big improvements to GPT-5.5 Instant, including being much more fun to talk to. Give it a try:

Agents are being adopted very quickly and accelerating work. How this looks across OpenAI itself: https://t.co/pQEiAUWuDa

Introducing Jalapeño — designed from scratch for LLM inference over nine months, accelerated by our models. Perf per watt looking incredible.

team cooked, spicily

you can now teach Codex by demonstration:

codex for testing every single feature in your app:

Codex Security plugin for security teams: deep scans, validating findings, tracing attack paths, building threat models, generating codebase-specific patches for review, and exporting into other tools: https://t.co/jxlKnsQOAw

digital ocean for running a codex remote session: https://t.co/IdCkX3QgOf

Launching credit usage analytics and updated spend controls for enterprises, available in our global admin console: https://t.co/fyT2JJ0FFl

OpenAI for Samsung:

We've collaborating with hundreds of physicians across 60 countries, 49 languages, and 26 specialties to make ChatGPT great at health-related questions for everyone:

OpenAI for helping families facing rare genetic diseases. This was done with o3 which is over a year old, amazing to think what will be possible with today’s models.

A super long overdue (3+ years?) post on scaling laws.

Compute is expensive. Scaling laws are a way to help us reason about the optimal compute allocation between data and model size before committing to a large run.

The post covers what scaling laws predict, how compute-optimal allocation works, why Kaplan et al. and Chinchilla disagree, and how data limits + fitting details make extrapolation tricky.

https://t.co/HP26eJvjHB

AI in 2040 will not be built on the stack we are using today. It will be much closer to optimal. The current stack has 3-4 orders of magnitude of data inefficiency and 4-5 orders of magnitude of compute inefficiency.

Near-optimal AI is what symbolic learning will deliver.

When the cost of execution drops, the value of taste, strategy, and architectural vision skyrockets. Previously, you were spending most of your cognitive budget on the micro. Now you are free to focus on the macro.

The market is treating Adobe like a legacy software company in terminal decline. Yet the actual data shows it's one of the biggest beneficiaries of the rise of GenAI. In fact, it's one of the top 5 most profitable & fastest-growing AI companies today, in an industry where profitability is rare.

Market efficiency is a myth. Markets are made of people, who are irrational both individually and in aggregate.

Maybe if human irrationality were completely random, it might average out to a rational whole. But in practice individual irrationality is driven by globally shared irrational narratives -- almost everybody is an irrational conformist.

If your benchmark relies on a static dataset or sampling from a static distribution densely known at training time, then it is fundamentally measuring memorization/retrieval. Which might be fine if you're looking for a retrieval benchmark! But don't confuse it with intelligence.

To be clear:

1. No it can't. I've used Fable while it was available, it was a good model but still less than 1% of the way there.

2. If it could, that fact would generally benefit SaaS companies, not hurt them. Those who stand to benefit most from better developer tools are software developers.

3. The code is not the product. No customer is going to want to take on every side quest they face simply because they can generate code for it, when they could just pay a subscription to get the problem out of their way.

4. Easier code means more software, which means more usage surface for existing SaaS. You're already seeing this trend in the data.

Maps are the primary force that shapes the territory they picture.

The most complex phenomena arise from scalable recombination of very simple rules. Whether it's galaxies, chips, or neural networks, if you find the right primitive building blocks, the complexity takes care of itself.

The best way to understand a complex system is via edge cases and failure modes, because they define the contour of the system.

The true measure of a software engineer isn't their ability to write clever code. It's their ability to ruthlessly protect the codebase from unnecessary cleverness.

Autonomy isn't the ability to act without human supervision. It's the ability to *learn* without human bottlenecks in the process. A system that is fully dependent on human training data and RL environments is only an imprint of human knowledge.

With agentic coding, complexity compounds in a mechanical way: unnecessary code ends up in the codebase, moves to the context window, degrades the model's reasoning abilities, leads to more unnecessary code (often to fix issues arising from the unnecessary code). It's exponential

It seems almost too dumb to be true, but apparently the literal belief of SaaS bears is "all software is a 0 because Claude can one-shot these apps"

Just staggering levels of short-sightedness in that statement

Agentic coding forces you to design clean interfaces and document them well. An agent cannot read the implicit mental model shared by your engineering team, it can only read your API contracts and docstrings.

Over the last two weeks, both the U.S. Government and Anthropic took significant actions that demonstrated their power to control access to AI by restricting what others can do with frontier models. This has been one of those moments that, once seen, will be hard to unsee, and it is significantly accelerating many businesses’ and nation states’ efforts to ensure reliable access to AI that no one else can terminate.

Anthropic first released Claude Fable 5, a version of its Mythos model with additional guardrails, including some restrictions that seem well justified on safety grounds (such as limitations on applying it to hacking, bioweapons, and so forth). However, it also restricted developers’ ability to use it to build competing LLM technology. This move was concerning, given that the whole AI community, including Anthropic, has benefitted tremendously from open research — indeed, the AI revolution was kicked off by my former team (Google Brain) freely publishing the Transformers paper!

Imagine if Microsoft’s terms of use barred anyone from using their tools to build competitive software, or if Google barred using it to search for information to work on competing search engines. Anthropic’s argument that it was unsafe for others to be able to make advances in AI also rang hollow. Initially, Anthropic silently degraded Fable 5’s performance for users detected to be working on LLM research through invisible interventions that weakened the model’s outputs without notifying the user. After significant backlash, it walked back this decision and decided to be transparent when it did this, but it still refuses to use its latest capabilities to help AI researchers.

This move represents a raw demonstration of power by Anthropic. It has used “safety” arguments to hinder potential competitors. Platforms succeed when they are viewed as stable, reliable partners that one can build on. The sudden rule changes by Anthropic (including a mandatory 30 day data retention policy for Fable usage) have made developers wonder about the stability of building on any one proprietary LLM provider, not just Anthropic.

The U.S. Government then shortly followed with an even greater demonstration of power. It used the Commerce Department’s authority to regulate technologies that may be national security threats to restrict exports of Mythos and Fable, requiring a license for use by any foreign national, whether inside or outside of the U.S., including employees of Anthropic. This led Anthropic to disable access to Fable to all users worldwide.

Sam Altman pointed out, referring to Anthropic, “It is clearly incredible marketing to say, ‘We have built a bomb, we are about to drop it on your head. We will sell you a bomb shelter for $100 million.’” But when one engages in this type of fear-based marketing, it increases the odds that the U.S. Government will agree with you and slap export controls on the bomb you say you have built.

To be clear, I don't think Anthropic has built anything like a bomb, and I don't think export controls on Fable are appropriate.

However, following the U.S. Government making this move, many nations, including U.S. allies, saw how the U.S. can suddenly yank their access to AI models. In many capitals around the world, this has spurred discussions on AI sovereignty and how others can ensure uninterrupted access to this critical technology.

For decades, many nations were comfortable having many parts of their supply chain rely on the U.S., China, and other major producers. Once a nation issues a threat, or takes action, to limit other nations’ access, other nations will rationally try to secure alternatives. For decades, semiconductor manufacturing in China made slow progress; once the U.S. moved to limit China’s access, China’s efforts kicked into high gear. Similarly, once China threatened U.S. access to rare earth minerals, U.S. efforts to secure alternatives accelerated. Now that it has become crystal clear that private U.S. companies and the U.S. government can limit, in short order, other nations’ access to frontier AI models, the incentive of others to invest more in alternatives like open source grows significantly. Of course, training frontier models is not easy, so it remains to be seen how successful they are, but we have crossed the rubicon.

Satya Nadella wrote an essay about the importance of building a healthy ecosystem on top of frontier AI technology. I heartily agree with him, and hope this week’s events will ultimately prove to be constructive steps toward this.

I hope we can build a more free, more open world, where research is freely shared, and laws and societal norms shape a level playing field that allows everyone to make progress. A silver lining of the events of these past two weeks is now that everyone better realizes key points of instability of the current system, we can all work to create a more stable foundation.

[Original text: The Batch newsletter]

Potential name for the AI industry regulatory authority:

AI Associated Institute of America, Inc or AIAIAI, pronounced “ay yai yai”

It's time for a national billionaires tax and a new social contract.

10% of Americans own two-thirds of the wealth. Wages have stagnated. The cost of living has skyrocketed. 

The system is fundamentally broken.

The federal tax code, a corporate code, and an inheritance code were written for a different set of Americans. 

It’s time for an economic reset.

https://t.co/nrRXzlFLKy

Full duration single-engine static fire test of Starship https://t.co/nfR8PvHpze

BREAKING: Starlink is providing free internet for Venezuela after devastating earthquakes.

• Free service through July 25 for people impacted by the devastating earthquakes in Venezuela.

• The free service applies to both new and existing customers.

• Starlink is also rapidly deploying terminals to the hardest-hit areas.

Elon Musk is once again standing with people in need.

Future of aircraft engineering https://t.co/woWiqyABLJ

FIGMA MOTION IS HERE
FIGMA MOTION IS HERE
FIGMA MOTION IS HERE

Live from Config 2026 https://t.co/ut7OFVZKQS

Nuestro tercer avión ya llegó a Venezuela y el cuarto ya va en camino.

Seguimos reforzando esta misión.

Dios bendiga a Venezuela 🇸🇻🇻🇪 https://t.co/dJiyNZ7I6S

Hemos encontrado con vida a Camila Sofía Medina Rivas, una niña de 15 años atrapada junto a su mascota en el noveno piso de un edificio colapsado.

Su madre la está esperando abajo con parte de nuestro equipo.

Aún debemos romper varias paredes para llegar hasta ella, por lo que nuestros equipos han llevado más herramientas.

Ya contamos con la maquinaria y el personal con la experiencia necesaria para abrir el paso.

Primero Dios, pronto lograremos rescatarla.

¡Hemos rescatado a Nayarit!

Ahora a luchar para estabilizarla.

Dios, permite que podamos salvarle la vida. https://t.co/4yAuovqW8R

‼️🚨🆘😭😭😭 Estos verdaderos héroes de #Venezuela, están dando su vida por salvar otras vidas. Sin cascos, sin implementos, con las uñas, dando una palabra de aliento. Tanta valentía y bondad por parte de algunos, me quedo con eso.
¡Dios bendiga a Venezuela! Esto es durísimo. https://t.co/eifjRMpdmF