Meta Researchers Show AI Agents Can Verify Code Without Running It — and Hit 93% Accuracy

Meta's semi-formal reasoning enables AI agents to verify code without executing it, achieving 93% accuracy. Implications for code review and RL training costs.