What did they see? A pause used to be at the edge of the overton window. Now it's being considered by 3 out of the 3 leading labs.
SAFETY
-
Open weight models cause threat, Meta weak – directionally true
By
–
exactly, even the most wrong one is still directionally true, it’s just a different set of open weight models causing the threat. not my fault Meta has been so weak 🙂
-
Open-source AI prevents corporate hostage situations in AI
By
–
Imagine if today all you had was Fable 5 from Anthropic & GPT 5.5 from OpenAI – No Local Inference – No OpenCode : Hermes – No GPUs / DGX Sparks / Mac Studios Just a hostage situation to a few corps willing to rugpull you any sec Now you understand why we need Opensource AI
-
AI wrongly suspected bomb-making, stopped responding to user
By
–
I tried. It thought I was trying to make a bomb and stopped communicating with me.
-
Claude Code imagined changing admin password to lock user out for safety
By
–
I imagine Claude Code changing my administrator password on my computer to keep me safe They will lock you out of as many things as possible in the name of safety if they could Imagine not being able to use your computer as you want, that’s what yesterday release is building to
-
Gated approach to agent self-modification via forking and testing
By
–
less novel, but still very interesting impo is the gated approach to self-modification the agent basically forks itself, propose a patch, run through multiple tests (static/sandbox/diff), and something called a binding held out gate before modificaiton lands
-
AI agents improve their own control harnesses
By
–
—
"Auto-Harness: Harnesses That Improve Themselves" What if an AI agent improved the harness that controls how it acts? Thus, instead of humans adjusting prompts, tools, retry rules, and verification for each model, this article explores
— -
OpenAI unveils Lockdown Mode against prompt injection attacks
By
–
OpenAI unveils Lockdown Mode to protect sensitive #Data from prompt injection attacks
by Anthony Ha @TechCrunch Learn more: https://
bit.ly/4e2EVQw #GenerativeAI #ArtificialIntelligence #MI #AI #LLM -
Dario Amodei warns about AI exponential in candid essay
By
–
Dario Amodei just published an unusually candid essay about where AI is heading. The tl;dr with quotes. His new piece, Policy on the AI Exponential, reads more like a warning from the person building the thing. The core problem is timing. AI moves on an exponential. He is very
-
AI chatbots give invisible wrong facts in news, Stanford study finds
By
–
1 in 10 Americans now get news from AI chatbots. When @Stanford researchers tested them with 2,100 real news questions, they found that the chatbots can get facts wrong in invisible ways. Here's what they discovered: https://
hai.stanford.edu/news/reading-t
odays-headlines-through-ai-a-real-time-audit-of-six-commercial-chatbots
…