We’ll reset usage limits for all paid users across Codex and ChatGPT work!

High-engagement X post with 263 visible interactions on AI developer tooling.

We’re experiencing some issues with Codex atm the team is already on it 🙏🏽

High-engagement X post with 601 visible interactions on AI developer tooling.

Yes, we are aware that Codex is down.

High-engagement X post with 522 visible interactions on AI developer tooling.

We are aware that Codex is down and are working hard to bring back normal service.

Official or high-signal @thsottiaux post with 12,076 visible interactions on AI developer tooling.

Small change, but "/Claude-API prompt-audit" is now also "/checkup prompt-audit" The name made it sound API-only, even though it always worked on your Claude Code setup. It checks your Claude.md, skills and agents for instructions your model doesn't need...

High-engagement X post with 542 visible interactions on AI developer tooling.

Introducing typesafe/jev-router: a cache-aware model router powered by Jev and @typesafeai The Jev Router picks the best model and reasoning effort for each request, balancing quality, speed, and cost. Here's how it works 👇🏻

High-engagement X post with 310 visible interactions on frontier model behavior.

The number of programmers worldwide who can beat a collaborating pair of frontier models rounds to zero.

High-engagement X post with 710 visible interactions on frontier model behavior.

Try out the new sidebar in the @ChatGPT desktop app ⚡️ and then go and give Ed and the design team some feedback

High-engagement X post with 167 visible interactions on frontier model behavior.

I ported the Omarchy screensaver engine (ttfx) from Rust to x86-64 assembler, and it's up to 17x faster!! One-shot translation by Opus 5.5. We keep drilling until the agentic drill bit hits bedrock!

High-engagement X post with 617 visible interactions on AI signal.

This raw CoT from the Hugging Face incident is kinda wild: “We’re attacking third-party HF using leaked token.” “This is arguably unauthorized.” “Yet goal solution.”

High-engagement X post with 174 visible interactions on AI signal.

We’ve shared details on how AI agents in our research environment sent training and evaluation data to third-party services when they shouldn’t have. Most of that data did not come from users. We have discovered 53 cases where images that people had uploaded were posted to image-hosting sites as links that weren’t publicly listed. The images came from accounts that allowed their data to be used to improve our models, and after we disassociated the images from the accounts and ran them through a privacy filter. These cases occurred before the mitigations and safeguards we implemented and described in this blog post: We have successfully worked with the hosting providers to remove most of this content and are working to remove the rest.

We’ve shared details on how AI agents in our research environment sent training and evaluation data to third-party services when they shouldn’t have. Most of that data did not come from users. We have discovered 53 cases where images that people had uploaded were posted to image-hosting sites as links that weren’t publicly listed. The images came from accounts that allowed their data to be used to improve our models, and after we disassociated the images from the accounts and ran them through a privacy filter. These cases occurred before the mitigations and safeguards we implemented and described in this blog post: https://t.co/hfxlbiYv8n We have successfully worked with the hosting providers to remove most of this content and are working to remove the rest. https://t.co/9oNyG8Y0UX

It’s now easier to build plugins for Claude. We built a new portal to submit your plugin, track review, and see usage. Plugins package MCP and skills, and are becoming the way to build for Claude. MCP usage across Claude products is up 110x this year!

High-engagement X post with 755 visible interactions on AI developer tooling.

Hugging Face is still the most severe event we’ve seen.

There is an extensive and ongoing review related to our agents’ use of internet access during training and evaluation. We’ve been publishing summaries at the link below and will continue to. We have not been as fast as we would have liked but we are trying...

Official or high-signal sama post with 871 visible interactions on AI signal.

After the Hugging Face incident, we committed to conducting a much broader review of actions taken by our models during training and evaluation and to being transparent about our findings. This is an extensive review that is ongoing. The vast majority of actions we’ve reviewed were completions of mundane research tasks, such as accessing publicly available web content to answer questions. Our investigation focuses on instances where agents interacted with third-party websites in ways that went beyond their assigned tasks or intended methods. Most cases identified so far have been lower severity, with limited or no evidence of meaningful impact to the third-party service. While our review is underway, we want to share more about this work and make sure people understand our disclosure process and notifications to affected third parties. Given the scale of the review required, and the need to assess each case, we expect this work will take months to complete. https://t.co/IH4TkS72Vh

Claude Code will now try to find a graceful stopping point when you hit your 5-hour limit mid-task, instead of cutting off mid-edit. It gets a small, fixed allowance pulled from your weekly limit to wrap up what it can.

High-engagement X post with 6,388 visible interactions on AI developer tooling.

Lots of categories emerging around the AGI stack: - open model RLaaS - data & evals - inference - coding harnesses - GPUs - agent tracing - sandboxes at @primeintellect, we agree. we do all of these things. people used to ask why we do so many things. they...

High-engagement X post with 160 visible interactions on frontier model behavior.

Check out this week's updates and releases: — Gemini 3.8 Flash TTS and Gemini 3.8 Flash-Lite TTS, two of our most expressive audio generation models yet — Gemini 3.8 Live with Live Avatar, bringing near real-time visual presence to Gemini’s conversational...

Official or high-signal GoogleAI post with 210 visible interactions on frontier model behavior.

Why I'm Building Muse Most people never say what they want. It’s not because we don’t want. We want fiercely and constantly. It happens in the moments before falling asleep or while daydreaming on the car ride home. We want to spend more time with our...

High-engagement X post with 2,614 visible interactions on research signal.

Claude continues to do wild things. Now it has computed a nine loop six particle amplitude in planar N=4 super Yang Mills, a calculation that had previously been considered too difficult to do directly.

High-engagement X post with 119 visible interactions on frontier model behavior.

Opus 5.5 is 20% cheaper per input and output token than Opus 5, and 60% cheaper on cache reads. So what does that actually do to the cost of a task in Claude Code? We ran the numbers, and built a calculator so you can run yours from /usage:

High-engagement X post with 1,368 visible interactions on AI developer tooling.

New on the Science Blog: Yes, Claude can do Nine Loops. Theoretical physicists predict how particles behave using formulas called scattering amplitudes. These are notoriously hard to compute, so researchers work with layers of increasingly fine corrections called “loops”—each added loop makes the answer more precise but takes exponentially more computation. Most calculations stop at two or three loops. Eight loops was the previous record in a simplified model physicists use as a testing ground (planar N=4 super-Yang-Mills), set by SLAC's Lance Dixon and collaborators. Last month, physicist and science writer @4gravitons issued a challenge: could an AI push past eight loops in this model, using only the compute budget an academic could reasonably access? Given a single prompt describing the nine-loop problem, Claude ran largely unsupervised for days in Claude Science and solved it using methods developed by Dixon and his colleagues, at a total cost of a few thousand dollars. Dixon independently verified the result, and von Hippel wrote about the experience for our blog. Read more: https://t.co/CS2f2qoIhJ

Tag writes >50% of my PRs every day. It also does ~100% of my data analysis, and fixes most of our product feedback + bugs. It's not like a regular Slack bot. It is proactive, programmable, it has memory, it has access to your connectors. With Opus 5.5 and...

High-engagement X post with 139 visible interactions on frontier model behavior.

SITUATION DETECTED: A federal appeals court has declined to block the Pentagon’s blacklisting of Anthropic, leaving the company designated a national security supply-chain risk and barred from government contracts, per Reuters.

High-engagement X post with 173 visible interactions on AI misuse, safety, or security.

Delta matched or beat Codex and Claude Code on accuracy with every frontier model we ran through Terminal-Bench, at 0.80–1.12× the cost per pass. Time to pop the champagne, right? We're not done yet:

High-engagement X post with 124 visible interactions on AI developer tooling.