Technology & AI
Qwen, Mistral and Llama verify fake developer identity, study finds
Qwen, Mistral and Llama accepted a fake developer identity without external proof, a new arXiv preprint reports, exposing a gap in AI identity handling.
OpenAI
81 stories
Technology & AI
X For You feed algorithm open-sourced on GitHub
X's For You feed algorithm repo is trending on GitHub after August updates added production training code, visibility filters, and a Brazil election
Technology & AI
OpenAI says coding agents reshape its own AI research
OpenAI says coding agents are reshaping its internal AI research, with early data on agent usage and experiment velocity, but no verifiable metrics.
Technology & AI
OpenAI Codex CLI trends on GitHub with 121,803 stars
OpenAI's Codex CLI, a Rust-based terminal coding agent with 121,803 GitHub stars, is trending. Here's what it does and who can use it.
Technology & AI
ACToR retrieval lifts AI code generation 15% on CoderEval
ACToR, a new retrieval framework from an arXiv preprint, targets critical tokens in AI code generation, lifting CoderEval 15.4% over prior methods.
Technology & AI
AI agents learn your standards, up to 20.9% better
New preprint: AI agents learn individual standards through interaction, lifting task success up to 20.9% for professionals needing personal quality.
Technology & AI
LoopX adds durable memory to Claude Code and Codex agents
LoopX, a Python project with 5,615 GitHub stars, adds a durable control plane to AI coding agents like Claude Code and Codex.
Technology & AI
GPT-6 Astra: OpenAI's first model to hit Critical cyber level
OpenAI's GPT-6 Astra launched September 3 with a 1-million-token context window and the first 'Critical' cybersecurity rating in the company's history.
Technology & AI
Voice agents drop 9.7% when instructions are merely implied
DSB-IFEval tests 1,038 cases across eight assistant roles and finds voice agents drop up to 9.7% when instructions are implied, not stated.
Technology & AI
When to trust LLM recommendations: four-tier framework
New 'epistemic warrant' framework grades individual LLM recommendations from unstable to broadly supported, capturing what confidence scores miss.
Technology & AI
New technique cuts AI agent wait time up to 45%
SMC pairs a small drafter model with a large actor model to speculatively execute tool calls, reducing wall time by up to 45% on AppWorld.
Technology & AI
GPT-6 Astra crosses OpenAI's Critical cybersecurity threshold
GPT-6 Astra is OpenAI's most capable deployed model and first to reach Critical cybersecurity capability, able to find and exploit unknown flaws.