AI Agents Just Crossed a Dangerous Line (SWE-bench 70%+) Channel: Best of Podcast8 views • 4 weeks agoRelated VideosBeyond SWE-Bench Pro - Where do Agents go from Here?SWE-Bench is getting replaced???METR: Half of SWE-Bench Passes Fail Real Code ReviewSWE Bench Verified - AI Benchmark