{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/constitutional-ai-harmlessness-from-ai","title":"Constitutional AI: Harmlessness from AI Feedback","arxiv_id":"2212.08073","date":"2022-12-15","proceeding":null,"authors":["Yuntao Bai","Saurav Kadavath","Sandipan Kundu","Amanda Askell","Jackson Kernion","Andy Jones","Anna Chen","Anna Goldie","Azalia Mirhoseini","Cameron McKinnon","Carol Chen","Catherine Olsson","Christopher Olah","Danny Hernandez","Dawn Drain","Deep Ganguli","Dustin Li","Eli Tran-Johnson","Ethan Perez","Jamie Kerr","Jared Mueller","Jeffrey Ladish","Joshua Landau","Kamal Ndousse","Kamile Lukosuite","Liane Lovitt","Michael Sellitto","Nelson Elhage","Nicholas Schiefer","Noemi Mercado","Nova DasSarma","Robert Lasenby","Robin Larson","Sam Ringer","Scott Johnston","Shauna Kravec","Sheer El Showk","Stanislav Fort","Tamera Lanham","Timothy Telleen-Lawton","Tom Conerly","Tom Henighan","Tristan Hume","Samuel R. Bowman","Zac Hatfield-Dodds","Ben Mann","Dario Amodei","Nicholas Joseph","Sam McCandlish","Tom Brown","Jared Kaplan"],"abstract":"As AI systems become more capable, we would like to enlist their help to supervise other AIs. We experiment with methods for training a harmless AI assistant through self-improvement, without any human labels identifying harmful outputs. The only human oversight is provided through a list of rules or principles, and so we refer to the method as 'Constitutional AI'. The process involves both a supervised learning and a reinforcement learning phase. In the supervised phase we sample from an initial model, then generate self-critiques and revisions, and then finetune the original model on revised responses. In the RL phase, we sample from the finetuned model, use a model to evaluate which of the two samples is better, and then train a preference model from this dataset of AI preferences. We then train with RL using the preference model as the reward signal, i.e. we use 'RL from AI Feedback' (RLAIF). As a result we are able to train a harmless but non-evasive AI assistant that engages with harmful queries by explaining its objections to them. Both the SL and RL methods can leverage chain-of-thought style reasoning to improve the human-judged performance and transparency of AI decision making. These methods make it possible to control AI behavior more precisely and with far fewer human labels.","url_abs":"https://arxiv.org/abs/2212.08073v1","url_pdf":"https://arxiv.org/pdf/2212.08073v1.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"constitutional-ai-harmlessness-from-ai","repo_url":"https://github.com/anthropics/constitutionalharmlessnesspaper","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok"}},{"paper_slug":"constitutional-ai-harmlessness-from-ai","repo_url":"https://github.com/uptrain-ai/uptrain","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok","spdx":"Apache-2.0"}}],"tasks":[{"task_slug":"decision-making","task_name":"Decision Making"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2212.08073","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2212.08073"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-24T18:15:14+00:00","read_at_is":"when the build read Syntology's graph, not when any sample ran","claim":"Per-sample execution status on synthesized fixtures; not a correctness claim about the paper. Samples come from repositories linked to the paper, official or community; repo_kind says which.","repos":[{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/uptrain-ai/uptrain","reach":{"status":"ok","spdx":"Apache-2.0"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/anthropics/constitutionalharmlessnesspaper","reach":{"status":"ok"}}],"summary":{"unverified":4},"by_repo_kind":{"listed":{"samples":4,"ran":0,"repositories":1}},"repo_kind_vocabulary":{"official":"The archive marks this repository official for the paper","named_in_paper":"The archive records that the paper mentions this repository; it is not marked official","listed":"In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper","found_in_text":"Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted","community":"Not in the archive's code links for this paper; a community repository Syntology harvested"},"n_pointer_only_for_licence":0,"samples":[{"code_sha256_prefix":"e8d89d69f3408074","entry":"format_uptrain_template","repo":"uptrain-ai/uptrain","repo_kind":"listed","path":"uptrain/integrations/promptfoo_utils.py","file_url":"https://github.com/uptrain-ai/uptrain/blob/HEAD/uptrain/integrations/promptfoo_utils.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"e8d89d69f3408074"}},{"code_sha256_prefix":"9e1d39d2fcccc91f","entry":"raise_or_return","repo":"uptrain-ai/uptrain","repo_kind":"listed","path":"uptrain/framework/remote.py","file_url":"https://github.com/uptrain-ai/uptrain/blob/HEAD/uptrain/framework/remote.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"9e1d39d2fcccc91f"}},{"code_sha256_prefix":"be851771a55d2bde","entry":"register_custom_op","repo":"uptrain-ai/uptrain","repo_kind":"listed","path":"uptrain/operators/base.py","file_url":"https://github.com/uptrain-ai/uptrain/blob/HEAD/uptrain/operators/base.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"be851771a55d2bde"}},{"code_sha256_prefix":"6f21bbd12e4d0294","entry":"register_op","repo":"uptrain-ai/uptrain","repo_kind":"listed","path":"uptrain/operators/base.py","file_url":"https://github.com/uptrain-ai/uptrain/blob/HEAD/uptrain/operators/base.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"6f21bbd12e4d0294"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}