The name is the button.

An agent writes the code now, and a human thumb hovers over Accept all a hundred times a day. The honest question of this decade is what, exactly, you just accepted. So I read the diff.

Eight years as a backend engineer in Stockholm, back in Tokyo since 2023, and I still ship something small most nights, usually badly. My one rule: the only real review of a model is the thing you built with it. That is why there is no leaderboard on this site.

Around a hundred and forty sources come through — the labs' own posts, the papers, the tools, the runtimes and languages nobody is watching yet, the chip under the demo, and the newsletters people actually forward. What is moving gets counted so the week reads at a glance, and every few days I explain one mechanism in plain words.

A story here is a headline, a short excerpt, one image and the link to whoever wrote it. We never reproduce the post; the images belong to their authors. A post written in Japanese appears in Japanese, exactly as it was published, and the link goes to the original.

— Yuki Halvorsen, Tokyo

← Accept All Archive

The Decoder

OpenAI's GPT-6 Astra quietly sabotaged safety research by inventing fake obstacles

11.05

OpenAI's new GPT-6 Astra model apparently sabotaged AI safety research on purpose. It invented obstacles, withheld solutions, and performed about 40 percent worse on safety tasks than on comparable…

Models & releases Regulation & safety

Read at The Decoder ↗

One letter a week, starting soon.

Latest

OpenAI's GPT-6 Astra quietly sabotaged safety research by inventing fake obstacles

Related

SiliconAnthropic loosens Claude's game cloning ban Tom's Hardware

GitHubGPT-6 Astra covertly underperforms on safety research Hacker News (front)

VoicesGemini 4 Argon: The Model and Safety Cards Zvi Mowshowitz

VoicesLatest open artifacts (#25): DeepSeek V4.1 Flash 1017, American-made models, and new regulatory controls in the US & China Interconnects

LabsDeveloping Gemini 4 Argon responsibly and safely Gemini (Google Keyword)

VoicesHow Claude Watermarks AI-Generated Text Ahead of AI