AI Safety Policy Needs to train Legal Practitioners
I completed my law degree at a working-class London university. In my first year, I was 18 years old, and I was often the youngest person…
Full feed
Feed 28,135 posts
Every story across every category, newest first. Each card links to the original publisher; daily-brief posts open as editorial pages.
Grok Build improves almost every day
Grok Build improves almost every dayJason Ginsberg: some special features in Grok Build if you're new/dashboard: shows you every agent running in your TUI. no need…
Grok gaming
Grok gamingtetsuo: Grok 4.5 in Grok Build created an FPS game in under an hour.The prompt was simple. I told it to write a game design…
Microsoft’s carbon emissions went up 25 percent last year
Microsoft may once again be struggling to keep up with its own climate goals, according to its 2026 sustainability report. As reported by GeekWire, the report…
Grok doesn’t give up
Grok doesn’t give upComposio: Grok 4.5 is the most persistent agent model we've tested. Here's one example: In one of our evals, we asked 3 models…
What would it take for AI to discover penicillin?
At the end of May, I attended the SciFM26 conference hosted at UChicago which had many speakers discussing agentic science and the future of AI+Science. This…
Getting started with ChatGPT
Learn how to use ChatGPT, start your first conversation, and discover simple ways to write, brainstorm, and solve problems with AI.
Behavioral Privacy Leakage in Agentic Negotiation: Formalizing and Mitigating Inference Attacks via Randomized Policies
This paper was accepted at the AI4TCI (Workshop on AI for Secure and Trustworthy Critical Infrastructure Systems) Workshop at the International Conference on Availability, Reliability and…
Working at the frontier: How Cognition trusts Claude Fable 5 to work through the night
Working at the frontier: How Cognition trusts Claude Fable 5 to work through the night
I like choices… but now I have: 2x modes (Codex vs. Work mode) 3x GPT-5.6 models (Sol, Terra, Luna) 5x effort levels (Light, Medium, High, Extra Hig…
I like choices... but now I have:2x modes (Codex vs. Work mode)3x GPT-5.6 models (Sol, Terra, Luna)5x effort levels (Light, Medium, High, Extra High, Ultra)That's 2…
Grok 4.5
Grok 4.5Kevin Bass: Grok 4.5 scores significantly better than other frontier models Opus 4.8 and GPT 5.5 across ALL professional work benchmarks.It performs exceptionally well at…
Has anyone tested how quantization hits different capabilities separately? My results are surprising.
I've been running some systematic tests on a few models comparing FP16 vs various GGUF quant levels, and instead of looking at one aggregate benchmark score,…
RT steve hsu: Research produced by LLMs tends to be combinatorial – a synthesis of ideas that already exist in the literature, rather than something t…
RT steve hsuResearch produced by LLMs tends to be combinatorial - a synthesis of ideas that already exist in the literature, rather than something truly original.…
Fidji Simo steps down from OpenAI’s no. 2 role
OpenAI's No. 2 executive, Fidji Simo, is stepping down from her full-time role after her medical leave proved longer than expected — a leadership vacuum that…
easy to overlook the implications for how Sol can accelerate your engineering workflows
easy to overlook the implications for how Sol can accelerate your engineering workflowsAndrew Curran: Quote from OpenAI on the livestream.'Already, Sol has been transforming our research…
RT Jeff Bezos: Wally Funk waited 60 years to get to space, and no one ever earned it more. She trained with the Mercury 13 in 1961, out-tested the men…
RT Jeff BezosWally Funk waited 60 years to get to space, and no one ever earned it more. She trained with the Mercury 13 in 1961,…
AI Daily Brief — 9 July 2026
OpenAI shipped GPT-5.6 across three price tiers and launched ChatGPT Work, while Meta released Muse Spark 1.1 and its first commercial API on the same day.
Hint for all AI Labs as they branch out from work for programming to general knowledge work: non-coders are not just dumber coders Taking away a bunch…
Hint for all AI Labs as they branch out from work for programming to general knowledge work: non-coders are not just dumber codersTaking away a bunch…
Fidji Simo steps down from leading OpenAI’s AGI work due to illness
OpenAI's Fidji Simo is departing her full-time role as the company's AGI chief and is transitioning to being a "part-time advisor," she said on X. The…
RT Jerry Liu: We comprehensively benchmarked GPT-5.6 on document understanding. At a high-level there's no change between GPT-5.6 Sol and GPT-5.5 in t…
RT Jerry LiuWe comprehensively benchmarked GPT-5.6 on document understanding.At a high-level there's no change between GPT-5.6 Sol and GPT-5.5 in terms of performance over tables, text,…
> if China could take Taiwan, a tiny island right next to it, with reasonable losses, they'd do it Man, burgers will be so shell-shocked if China "tak…
> if China could take Taiwan, a tiny island right next to it, with reasonable losses, they'd do itMan, burgers will be so shell-shocked if China…
RoboTALES: Learning Reasoning-Guided Robot Policies via Task-Aligned Simulated Futures
Pretrained video generative models are promising backbones for visuomotor control, but their imagined futures often drift from task intent and are not reliably action-conditional. As a…
i am really sad about this and very grateful for all fidji has done for openai, and even grateful for her friendship and who she is as a person. we al…
i am really sad about this and very grateful for all fidji has done for openai, and even grateful for her friendship and who she is…
RT Sarah Chieng: Technical writing is the #1 top-of-funnel motion at a $12B company @philipkiely broke down exactly how he does it, his full writing p…
RT Sarah ChiengTechnical writing is the #1 top-of-funnel motion at a $12B company@philipkiely broke down exactly how he does it, his full writing process, the hindsight…