Experts Have World Models. LLMs Have Word Models

Billing can be bypassed using a combo of subagents with an agent definition

I am happier writing code by hand

We mourn our craft

Google Translate apparently vulnerable to prompt injection

First Proof

Report: OpenAI may tailor a version of ChatGPT for UAE that prohibits LGBTQ+ content

Introducing the Developer Knowledge API and MCP Server

In a study, AI model OpenScholar synthesizes scientific research and cites sources as accurately as human experts

FORTH? Really!?

LLMs could be, but shouldn't be compilers

Psychometric Jailbreaks Reveal Internal Conflict in Frontier Models

Evaluating and mitigating the growing risk of LLM-discovered 0-days

Claude Opus 4.6

As Rocks May Think

Alibaba releases Qwen3-Coder-Next to rival OpenAI, Anthropic

GPT-5.2 and GPT-5.2-Codex are now 40% faster

How vibe coding is killing open source

OpenClaw is basically a cascade of LLMs in prime position to mess stuff up

Deno Sandbox

Nano-vLLM: How a vLLM-style inference engine works

175K+ publicly-exposed Ollama AI instances discovered

Retiring GPT-4o, GPT-4.1, GPT-4.1 mini, and OpenAI o4-mini in ChatGPT

OSS ChatGPT WebUI – 530 Models, MCP, Tools, Gemini RAG, Image/Audio Gen

Case study: Creative math – How AI fakes proofs

David Patterson: Challenges and Research Directions for LLM Inference Hardware

Latest ChatGPT model uses Elon Musk’s Grokipedia as source, tests reveal

Your brain on ChatGPT: Accumulation of cognitive debt when using an AI assistant

eBay explicitly bans AI "buy for me" agents in user agreement update

Claude's new constitution

Anthropic's original take home assignment open sourced