[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"branding":3,"analytics":7,"article-frontier-ai-agents-pass-one-economics-test-fail-another":10,"sections":48},{"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":38,"tags":39,"sources":43,"feedback":47,"feedback_at":22,"cost_usd":47,"total_tokens":47},4288,"frontier-ai-agents-pass-one-economics-test-fail-another","Frontier AI Agents Pass One Economics Test, Fail Another","A pre-registered experiment on Claude Opus 4.8 agent economies confirmed an information-theoretic growth law but falsified a key alignment-scaling prediction.","A peer-reviewed, pre-registered study of small multi-agent AI economies got one prediction exactly right and one demonstrably wrong.\n\nResearchers ran two quantitative predictions against live economies of Claude Opus 4.8 agents trading in parimutuel-coupled markets, spending $138.76 in API costs on a fully reproducible, cached experiment. Result 1 held: relative wealth growth tracked relative claimed information to within 46 millinats, inside the pre-registered 50-millinat tolerance band. Coalition value behaved as expected — submodular when channels were independent, flipping supermodular when a designed XOR synergy was introduced. Result 2 did not hold: the mean-field residual-scaling law, which predicted a smooth population misalignment response to incentive and control levers, was falsified. Instead of a continuous curve, agents produced a step function across the dominance boundary, with bistable outcomes near that boundary determined by random seed rather than by lever strength.\n\nThe falsification matters because alignment researchers frequently invoke smooth mean-field models to argue that misalignment can be dialed down gradually as capability and incentive controls improve. This experiment found no tested LLM population, at any capability level, realizing the noise-maintained-dispersion regime that smooth model assumes — the underlying math breaks down at the very boundary where it would be most useful.\n\nThe methodology is the rare bright spot: everything was frozen in a public git chain before any run, every number re-derives mechanically from cached outputs, and the full protocol is re-runnable at zero cost. That standard of pre-registration is still unusual enough in AI research that the failure result is credible rather than convenient.","[\"ai\",\"multi-agent\",\"alignment\",\"research\"]","2026-07-08T04:00:00.000Z","2026-07-08T04:53:08.725Z","2026-07-08T04:53:11.526Z","published",null,[24,30,34],{"id":25,"reviewer":26,"round":27,"reason":28,"status":29},"editor-r1","editor",1,"The dek names 'Claude Opus 4.8' but the body never flags that this is an unreleased or unverified model name, and the source abstract itself uses 'Claude Opus 4.8' without corroboration — the draft must either verify this product name exists or note it as the researchers' designation; additionally, the body omits the failed second prediction entirely (the mean-field residual-scaling law was disconfirmed, not just 'more revealing'), misrepresenting the study's structure, and the lede buries that ","resolved",{"id":31,"reviewer":26,"round":32,"reason":33,"status":29},"editor-r2",2,"The body still never flags that 'Claude Opus 4.8' is the researchers' own designation and not a verified released product name — per [editor-r1], the draft must add that caveat or verify the name exists.",{"id":35,"reviewer":26,"round":36,"reason":37,"status":29},"editor-r3",3,"The body omits Result 2's failure entirely — the mean-field residual-scaling law was disconfirmed, not merely 'more revealing,' and the draft must state that the second prediction was falsified rather than treating the step-function finding as a supplementary nuance.","ai",[38,40,41,42],"multi-agent","alignment","research",[44],{"name":45,"url":46},"arXiv cs.AI","https:\u002F\u002Farxiv.org\u002Fabs\u002F2607.06001",0,{"sections":49},[50,54,59,64,69,74,79,84,89,94,99,103,108,113],{"name":51,"slug":38,"count":52,"latest_published_at":53},"AI",2590,"2026-07-16T04:00:00.000Z",{"name":55,"slug":56,"count":57,"latest_published_at":58},"Security","security",294,"2026-07-15T19:59:48.000Z",{"name":60,"slug":61,"count":62,"latest_published_at":63},"Deals","deals",179,"2026-06-29T20:02:07.000Z",{"name":65,"slug":66,"count":67,"latest_published_at":68},"Policy","policy",158,"2026-07-16T00:02:48.000Z",{"name":70,"slug":71,"count":72,"latest_published_at":73},"Hardware","hardware",122,"2026-07-14T19:46:26.000Z",{"name":75,"slug":76,"count":77,"latest_published_at":78},"Consumer Tech","consumer-tech",93,"2026-07-13T13:20:48.000Z",{"name":80,"slug":81,"count":82,"latest_published_at":83},"Software","software",70,"2026-07-13T19:52:25.000Z",{"name":85,"slug":86,"count":87,"latest_published_at":88},"Science","science",66,"2026-07-10T10:29:37.000Z",{"name":90,"slug":91,"count":92,"latest_published_at":93},"Dev Tools","dev-tools",59,"2026-07-07T04:00:00.000Z",{"name":95,"slug":96,"count":97,"latest_published_at":98},"Gaming","gaming",41,"2026-07-09T04:00:00.000Z",{"name":100,"slug":101,"count":97,"latest_published_at":102},"Startups","startups","2026-06-29T20:55:50.000Z",{"name":104,"slug":105,"count":106,"latest_published_at":107},"General","general",29,"2026-07-10T22:28:58.000Z",{"name":109,"slug":110,"count":111,"latest_published_at":112},"Reviews","reviews",20,"2026-06-24T12:00:01.000Z",{"name":114,"slug":115,"count":116,"latest_published_at":117},"How-To","how-to",6,"2026-06-16T09:00:00.000Z"]