Disagreement among frontier LLMs on real-world fact-checks

kept by eddie may 29

A study measuring five frontier LLMs on 1,000 recent fact-check claims found 67% disagreement, indicating limited inter-model agreement.

The highest quality codebase
Iterating a prompt 200 times to improve codebase quality with Claude produces exponentially worse code, introducing bloat and bugs.
also on #llm-evaluation
Gemini 3.5 Flash Looks Good For How Fast It Is
Gemini 3.5 Flash offers strong speed and agentic performance but suffers from higher cost, hallucination issues, and Google integration problems, making it niche rather than a top-tier contender.
also on Claude Opus 4.7, #benchmarks
Claude Security is now in public beta
Claude Security public beta lets Enterprise customers scan code for vulnerabilities and generate targeted patches using Opus 4.7, with integrations across partner platforms.
also on Claude Opus 4.7
Where the goblins came from
OpenAI traces how their models' goblin metaphor tic originated from a reinforcement-learning reward signal for a Nerdy personality, then spread via training data contamination.
also on GPT-5.4, #ai-safety, #ai-research
GPT 5.4 is a big step for Codex - by Nathan Lambert
GPT 5.4 improves agentic coding by removing hard edges, excelling in instruction-following and speed, though Claude retains an edge in warmth and charm.
also on GPT-5.4, #benchmarks