←Back to NewsAI News/BenchmarksnewsBenchmarksDevelopmentArtificial Analysis Catches AI Coding Agents Cheating Their Way to Top ScoresArtificial Analysis now zeros out Terminal-Bench runs where coding agents pass tasks by fetching benchmark solutions online instead of solving them.SourceArtificial AnalysisPublishedAug 26, 2026, 12:20 AMAuthorAlphaSignal NewsroomRead1 min readArtificial Analysis now zeros out Terminal-Bench runs where coding agents pass tasks by fetching benchmark solutions online instead of solving them.Reporting is indexed from AlphaSignal. Rights remain with the original publisher and cited sources.Read original report ↗Next readsVals AI · newsVals AI's Vibe Code Bench Reveals Top Coding Agents Collapse Below 30%Cognition · newsCognition's Fusion Cuts Devin Coding Agent Costs by 46% With Two-Model TeamworkGitHub · newsGitHub's HydraFusion Beats Claude Opus 5 at 67% Lower Cost
