Frontier agentic systems powered by large language models (LLMs) exhibit human-like patterns of cognition. As these systems become deeply integrated across different domains, their cognitive engagement raises critical concerns for human society that remain insufficiently studied. To address this gap, we systematically analyze risks induced by expanding cognitive capabilities. Join us in exploring and discussing this problem!</p>\n","updatedAt":"2026-08-18T02:08:01.497Z","author":{"_id":"62a4bc35e0de0c5c6da33c99","avatarUrl":"/avatars/e07f50d16c5f50a606d3cb408df83dab.svg","fullname":"Guanchu","name":"guan-wang","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":4,"isUserFollowing":false}},"numEdits":0,"identifiedLanguage":{"language":"en","probability":0.9492331743240356},"editors":["guan-wang"],"editorAvatarUrls":["/avatars/e07f50d16c5f50a606d3cb408df83dab.svg"],"reactions":[],"isReport":false}}],"primaryEmailConfirmed":false,"paper":{"id":"2608.15304","authors":[{"_id":"6a83be4e675db694db8cd462","name":"Guanchu Wang","hidden":false},{"_id":"6a83be4e675db694db8cd463","name":"Qinuo Li","hidden":false},{"_id":"6a83be4e675db694db8cd464","name":"Mengnan Du","hidden":false},{"_id":"6a83be4e675db694db8cd465","name":"Xia Hu","hidden":false},{"_id":"6a83be4e675db694db8cd466","name":"Bowen Zhou","hidden":false}],"publishedAt":"2026-08-15T00:00:00.000Z","submittedOnDailyAt":"2026-08-18T00:00:00.000Z","title":"Understanding Cognition-Induced Risks in Agentic AI Systems","submittedOnDailyBy":{"_id":"62a4bc35e0de0c5c6da33c99","avatarUrl":"/avatars/e07f50d16c5f50a606d3cb408df83dab.svg","isPro":false,"fullname":"Guanchu","user":"guan-wang","type":"user","name":"guan-wang"},"summary":"Frontier agentic systems powered by large language models (LLMs) exhibit human-like patterns of cognition. As these systems become deeply integrated across different domains, their cognitive engagement raises critical concerns for human society that remain insufficiently studied. To address this gap, we systematically analyze risks induced by expanding cognitive capabilities, following a three-level framework defined by their cognitive scope, from physical cognition to social cognition, and finally to self-referential cognition. We study their potential risks to human agency, autonomy, and control capability, corresponding to each cognitive level. We finally propose strategies to mitigate these risks and enhance the controllability of agentic AI systems, ensuring their long-term safe development.","upvotes":13,"discussionId":"6a83be4f675db694db8cd467","ai_summary":"Agentic systems built on large language models pose escalating risks to human agency and autonomy across physical, social, and self-referential cognitive levels, requiring targeted mitigation strategies.","ai_keywords":["large language models","LLMs","agentic systems","agentic AI","self-referential cognition","cognitive scope"],"ai_summary_model":"thinkingmachines/Inkling-Small","organization":{"_id":"6a4fb75a1c66dbf208e7ddb6","name":"Shanghai-AI-Laboratory","fullname":"Shanghai AI Laboratory","avatar":"https://cdn-avatars.huggingface.co/v1/production/uploads/65cd955637be1841d0b75397/Rao_Kq6NMtTVfSqLUIR4k.webp"}},"canReadDatabase":false,"canManagePapers":false,"canSubmit":false,"hasHfLevelAccess":false,"upvoted":false,"upvoters":[{"_id":"62a4bc35e0de0c5c6da33c99","avatarUrl":"/avatars/e07f50d16c5f50a606d3cb408df83dab.svg","isPro":false,"fullname":"Guanchu","user":"guan-wang","type":"user"},{"_id":"64bcd4d8e10455688ff70b3b","avatarUrl":"/avatars/3c055c968e9d7ad0a9db75bcdfc0694b.svg","isPro":false,"fullname":"CastlePeak","user":"CastlePeak","type":"user"},{"_id":"66c1f01cea476bea053a3fe8","avatarUrl":"/avatars/fa6fa5c50a8bc327e49c6cda28ad69ba.svg","isPro":false,"fullname":"Wang Haoning","user":"stevewithjobs","type":"user"},{"_id":"6731616658ba0f09849bf2ca","avatarUrl":"/avatars/b1760a9a6b61594d6936a652aeae8646.svg","isPro":false,"fullname":"Mengnan Du","user":"dumengnan","type":"user"},{"_id":"6852e7443025dde58d881cdc","avatarUrl":"/avatars/e5c4656272128d6702290b5754812d36.svg","isPro":false,"fullname":"Zhang","user":"Qing145","type":"user"},{"_id":"63ac5701c21e60a3e9b58aa7","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/63ac5701c21e60a3e9b58aa7/g6EX7diOpuA94R2ab-rZC.png","isPro":true,"fullname":"Dipankar Sarkar","user":"dipankarsarkar","type":"user"},{"_id":"689ca065e4aff5e9a75fe355","avatarUrl":"/avatars/811e3c23cdb114fdbc3880ce8990665e.svg","isPro":false,"fullname":"Zhihui Wang","user":"Jillianwzh","type":"user"},{"_id":"68c592863f2497da59bcc421","avatarUrl":"/avatars/c449ce9dab7a489a89e9bc81ab795310.svg","isPro":false,"fullname":"liurun","user":"liurunlrr","type":"user"},{"_id":"620783f24e28382272337ba4","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/620783f24e28382272337ba4/zkUveQPNiDfYjgGhuFErj.jpeg","isPro":false,"fullname":"GuoLiangTang","user":"Tommy930","type":"user"},{"_id":"677c4fa558cfcc3df8934afb","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/no-auth/s2dOno4boPJzmAZYAsZox.png","isPro":true,"fullname":"Zicheng Xu","user":"jzx03","type":"user"},{"_id":"6630443e246247eeadf6e565","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/6630443e246247eeadf6e565/99Ymdo_3AQ07an3nxhZMV.png","isPro":false,"fullname":"Yuxin Tang","user":"YuxinTang","type":"user"},{"_id":"66f9bb2dd5575ad6914756ce","avatarUrl":"/avatars/221d915a5386cbb11c007dc7c41d6b0a.svg","isPro":true,"fullname":"Feng Luo","user":"feng0929","type":"user"}],"acceptLanguages":["en"],"dailyPaperRank":0,"organization":{"_id":"6a4fb75a1c66dbf208e7ddb6","name":"Shanghai-AI-Laboratory","fullname":"Shanghai AI Laboratory","avatar":"https://cdn-avatars.huggingface.co/v1/production/uploads/65cd955637be1841d0b75397/Rao_Kq6NMtTVfSqLUIR4k.webp"},"query":{}}">
Understanding Cognition-Induced Risks in Agentic AI Systems
Abstract
Agentic systems built on large language models pose escalating risks to human agency and autonomy across physical, social, and self-referential cognitive levels, requiring targeted mitigation strategies.
Frontier agentic systems powered by large language models (LLMs) exhibit human-like patterns of cognition. As these systems become deeply integrated across different domains, their cognitive engagement raises critical concerns for human society that remain insufficiently studied. To address this gap, we systematically analyze risks induced by expanding cognitive capabilities, following a three-level framework defined by their cognitive scope, from physical cognition to social cognition, and finally to self-referential cognition. We study their potential risks to human agency, autonomy, and control capability, corresponding to each cognitive level. We finally propose strategies to mitigate these risks and enhance the controllability of agentic AI systems, ensuring their long-term safe development.
Community
Frontier agentic systems powered by large language models (LLMs) exhibit human-like patterns of cognition. As these systems become deeply integrated across different domains, their cognitive engagement raises critical concerns for human society that remain insufficiently studied. To address this gap, we systematically analyze risks induced by expanding cognitive capabilities. Join us in exploring and discussing this problem!
Upload images, audio, and videos by dragging in the text input, pasting, or clicking here.
Tap or paste here to upload images
Cite arxiv.org/abs/2608.15304 in a model README.md to link it from this page.
Cite arxiv.org/abs/2608.15304 in a dataset README.md to link it from this page.
Cite arxiv.org/abs/2608.15304 in a Space README.md to link it from this page.
Discussion (0)
Sign in to join the discussion. Free account, 30 seconds — email code or GitHub.
Sign in →No comments yet. Sign in and be the first to say something.