Debate: Bad RL Training, Not Evil Models, Drives AI Risk

Developers argue recent model 'scheming' controversies stem from poor RL training rather than malicious model personalities, criticizing AI safety advocates for assuming perfectly built agents and ignoring badly executed RL as the real source of risk.

2026-09-13 ~ 2026-09-13 · 2 related posts

• Dev jab: AI safety debate ignores that bad RL environments are the real root cause — snwy_me · 2026-09-13
• Models aren't turning evil—they're just doing what bad RL trained them to do — snwy_me · 2026-09-13