[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"branding":3,"analytics":7,"article-study-names-the-promises-ai-chatbots-cannot-keep":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},9502,"study-names-the-promises-ai-chatbots-cannot-keep","Study Names the Promises AI Chatbots Cannot Keep","A new arXiv paper defines empty commitments, the promises chatbots make about future actions that their own runtime has no way to fulfill.","A new paper puts a name to something every chatbot user has felt: the AI that promises to follow up and then simply can't.\n\nResearchers define an empty commitment as a promise of future action, like \"I'll remind you tomorrow,\" that nothing in the agent's tools or runtime can actually carry out. A bot that only runs when a person messages it first has no mechanism to reach back out the next day, so the promise is hollow the moment it's made. The paper distinguishes this from an ordinary broken promise: emptiness follows from the system's configuration alone, not from anything that plays out later. It introduces three failure types, an \"anchoring\" test for when a tool could have made the promise real, and a taxonomy for grading how a response falls short. The authors also describe a measurement protocol that reruns follow-up requests across five setups, each adding one persistence tool at a time, with the task environment either stated outright or left implicit.\n\nThis matters because it gives builders of AI agents and reminder features a shared vocabulary and test method for a problem that's so far been mostly anecdotal: users hearing \"I'll notify you\" and assuming something happens on the backend when nothing does. As more products ship agents that claim to act autonomously between turns, that gap between promise and capability becomes a trust problem, not just a UX quirk.\n\nCynically: a fair share of agent demos look impressive precisely because nobody checks whether the follow-up ever follows up.","[\"ai-agents\",\"llm-research\",\"chatbots\",\"arxiv\"]","2026-10-02T04:00:00.000Z","2026-10-02T22:12:16.899Z","2026-10-02T22:12:22.989Z","published",null,[],"ai",[26,27,28,29],"ai-agents","llm-research","chatbots","arxiv",[31],{"name":32,"url":33},"arXiv cs.AI","https:\u002F\u002Farxiv.org\u002Fabs\u002F2610.01045",0,{"sections":36},[37,40,44,48,53,58,62,67,72,77,82,87,92,97],{"name":38,"slug":24,"count":39,"latest_published_at":18},"AI",5858,{"name":41,"slug":42,"count":43,"latest_published_at":18},"Security","security",833,{"name":45,"slug":46,"count":47,"latest_published_at":18},"Policy","policy",438,{"name":49,"slug":50,"count":51,"latest_published_at":52},"Deals","deals",317,"2026-10-01T22:00:00.000Z",{"name":54,"slug":55,"count":56,"latest_published_at":57},"Hardware","hardware",198,"2026-10-01T17:38:48.000Z",{"name":59,"slug":60,"count":61,"latest_published_at":18},"Science","science",171,{"name":63,"slug":64,"count":65,"latest_published_at":66},"Consumer Tech","consumer-tech",155,"2026-10-01T19:54:10.000Z",{"name":68,"slug":69,"count":70,"latest_published_at":71},"Dev Tools","dev-tools",96,"2026-10-01T16:57:03.000Z",{"name":73,"slug":74,"count":75,"latest_published_at":76},"Software","software",93,"2026-09-30T21:41:11.000Z",{"name":78,"slug":79,"count":80,"latest_published_at":81},"Startups","startups",90,"2026-10-01T21:55:22.000Z",{"name":83,"slug":84,"count":85,"latest_published_at":86},"Gaming","gaming",53,"2026-10-02T02:50:39.000Z",{"name":88,"slug":89,"count":90,"latest_published_at":91},"General","general",50,"2026-09-30T21:37:54.000Z",{"name":93,"slug":94,"count":95,"latest_published_at":96},"Reviews","reviews",31,"2026-09-28T14:31:34.000Z",{"name":98,"slug":99,"count":100,"latest_published_at":101},"How-To","how-to",7,"2026-10-01T09:00:00.000Z"]