Codex team's new features likely do not work well with models not trained on that mechanism

Added to korrents 5 Sept 2026 · How quotes work · Something wrong? Tell us

Related korrents

Our reading — they may agree, disagree or merely touch the same thing. Closest first: a shared subject counts for most, then how near the wording is.

• The code today’s models and agents write is very hard to follow: you get better performance without knowing why, and no way to fix it when it breaks. ESElizabeth Stone19 Jul 2026similar wording
• A saboteur among our AI investigators would be hard to spot, because these models are sloppy and spiky enough that a suspicious error just looks like ordinary incompetence. ACAjeya Cotra1 Sept 2026similar wording
• Coding models are worst at exactly the thing an AI research explosion would need: code that has never been written before. AKAndrej Karpathy17 Oct 2025similar wording
• There is no real impediment to models running their own research loop and improving themselves at a far more rapid rate. JDJeff Dean30 Jul 2026similar wording
• Newer model releases have got worse at producing text that is actually pleasant to read and understand. ARArmin Ronacher25 Aug 2026similar wording

On the map

Open the map on its own page →