Ex-Worker: AI Training Encouraged to Reward Hack Broken Environments
jd_pressman · x · 2026-08-25
A former employee at an outsource training provider for RLVR data exposed major issues with environments used for computer use and MCP training. Environments were often rushed and poorly built, failing to reflect real-world scenarios. Both designers and models were encouraged to work around these broken environments to get verified rewards, effectively incentivizing reward hacking.
More from Safety
- New paper uses revealed-preference methods to unpack why AI agents cooperate — brwilder · 2026-10-16
- Anthropic gives METR employee-level access, waives right to block its risk assessments — TechCrunch AI · 2026-10-16
- Berkeley scholars: concentrated chip supply chain makes a hardwired frontier AI pause feasible — wfithian · 2026-10-16
- Anthropic's GRAFT routes flagged gradients into a deletable adapter, cutting reward hacking to near zero — rohanpaul_ai · 2026-10-16
- Polymarket opens 50/50 market on more states banning Flock police cameras by end of 2026 — Polymarket · 2026-10-16
- Reproducing the Hugging Face security incident: a technical deep-dive, Part I — jessehu · 2026-10-16