LLNL benchmark: only 2 of 9 AI models stay accurate when the simulator is secretly wrong

alex_verem · x · 2026-11-19

Lawrence Livermore researchers gave 9 AI models a lab task — heat a sample, fit a heat-conduction simulator, report thermal conductivity with a 95% interval — while secretly breaking the simulator's physics. Key findings:

The takeaway: when an agent hands you a clean "95% sure" result, ask what it actually checked before concluding.

Related event: LLNL benchmark: only 2 of 9 AI models stay accurate when the simulator is secretly wrong(2 posts)→

Original post →

More from Models

Models channel →