[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"branding":3,"analytics":7,"article-why-throwing-more-compute-at-ai-agents-stops-helping":10,"sections":35},{"siteName":4,"siteTagline":5,"publisherName":4,"contactEmail":6},"The Revision","Tech news, decoded.","editor@therevision.news",{"gaMeasurementId":8,"adsenseClientId":9},"G-ZW2MV82GYR","ca-pub-8533917693782264",{"article":11},{"id":12,"slug":13,"title":14,"dek":15,"body_md":16,"tags_json":17,"published_at":18,"created_at":19,"updated_at":20,"status":21,"review_note":22,"review_notes":23,"image_url":22,"persona_id":22,"persona_name":22,"section":24,"tags":25,"sources":30,"feedback":34,"feedback_at":22,"cost_usd":34,"total_tokens":34},8815,"why-throwing-more-compute-at-ai-agents-stops-helping","Why Throwing More Compute at AI Agents Stops Helping","A new benchmark shows ten leading AI agents plateau on realistic tasks no matter how much extra compute or extra tries you give them.","Giving an AI agent more time to think, or more attempts at a task, does not reliably make it better at real jobs, according to a new study.\n\nResearchers built a new benchmark, General AgentBench, that mixes search, coding, reasoning, and tool-use tasks into one realistic test instead of grading agents on narrow, single-skill benchmarks. They ran ten leading AI agents through it and found scores dropped sharply compared to how those same systems perform on the specialized tests vendors usually cite. The team then tried two common fixes: letting an agent keep working longer on a problem, known as sequential scaling, and having it generate many attempts at once and pick the best one, known as parallel scaling. Neither approach produced consistent gains once the tasks got realistic.\n\nThe industry's default answer to weak AI performance has been more compute: more reasoning steps, more retries, more sampled attempts. This study argues that trick runs out of road fast in messy, real-world tasks, for two reasons. Agents hit a performance ceiling no matter how long they keep working, and when they generate multiple attempts, nothing reliably tells them which one is actually correct.\n\nThat second problem, the inability to verify its own answers, looks like the bigger weakness. It's a quiet rebuttal to two years of pitches that insist the fix for a weak model is just letting it think longer.","[\"ai agents\",\"benchmarks\",\"test-time compute\",\"llm\"]","2026-09-30T04:00:00.000Z","2026-10-01T04:41:18.919Z","2026-10-01T04:41:23.786Z","published",null,[],"ai",[26,27,28,29],"ai agents","benchmarks","test-time compute","llm",[31],{"name":32,"url":33},"arXiv cs.AI","https:\u002F\u002Farxiv.org\u002Fabs\u002F2602.18998",0,{"sections":36},[37,41,46,51,56,61,66,71,76,80,85,90,95,100],{"name":38,"slug":24,"count":39,"latest_published_at":40},"AI",5236,"2026-09-30T20:13:50.000Z",{"name":42,"slug":43,"count":44,"latest_published_at":45},"Security","security",799,"2026-09-30T19:29:08.000Z",{"name":47,"slug":48,"count":49,"latest_published_at":50},"Policy","policy",426,"2026-09-30T19:46:31.000Z",{"name":52,"slug":53,"count":54,"latest_published_at":55},"Deals","deals",298,"2026-09-30T21:00:26.000Z",{"name":57,"slug":58,"count":59,"latest_published_at":60},"Hardware","hardware",196,"2026-09-30T13:00:00.000Z",{"name":62,"slug":63,"count":64,"latest_published_at":65},"Science","science",157,"2026-09-30T15:00:56.000Z",{"name":67,"slug":68,"count":69,"latest_published_at":70},"Consumer Tech","consumer-tech",148,"2026-09-30T17:53:35.000Z",{"name":72,"slug":73,"count":74,"latest_published_at":75},"Dev Tools","dev-tools",92,"2026-09-30T18:17:45.000Z",{"name":77,"slug":78,"count":74,"latest_published_at":79},"Software","software","2026-09-30T17:45:00.000Z",{"name":81,"slug":82,"count":83,"latest_published_at":84},"Startups","startups",83,"2026-09-29T21:51:36.000Z",{"name":86,"slug":87,"count":88,"latest_published_at":89},"Gaming","gaming",50,"2026-09-30T16:24:30.000Z",{"name":91,"slug":92,"count":93,"latest_published_at":94},"General","general",49,"2026-09-28T16:44:57.000Z",{"name":96,"slug":97,"count":98,"latest_published_at":99},"Reviews","reviews",31,"2026-09-28T14:31:34.000Z",{"name":101,"slug":102,"count":103,"latest_published_at":104},"How-To","how-to",6,"2026-06-16T09:00:00.000Z"]