{"about":{"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.","site":"https://codewithpapers.app","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","syntology":{"site":"https://syntology.ai","developers":"https://syntology.ai/developers","mcp":{"server":"https://syntology.ai/mcp","transport":"streamable-http","server_card":"https://syntology.ai/.well-known/mcp/server-card.json","auth":{"type":"trial token, no account","trial_token":"https://syntology.ai/api/oauth/trial/token","method":"POST","docs":"https://syntology.ai/developers"}},"have":"https://syntology.ai/api/graph/have?x=<method, arXiv id or title> (free, answers coverage only)","paper_base":"https://syntology.ai/paper/","atlas_base":"https://app.syntology.ai/?focus="},"machine_readable":[{"url":"https://codewithpapers.app/llms.txt","what":"the machine catalog: every machine-readable file, counted"},{"url":"https://codewithpapers.app/index/manifest.json","what":"paper-to-code index by arXiv id, with Syntology's counts"},{"url":"https://codewithpapers.app/search/manifest.json","what":"site search index (titles, authors) and its files"},{"url":"https://codewithpapers.app/download","what":"bulk files: Syntology's layer, described there"},{"url":"https://codewithpapers.app/build_manifest.json","what":"the build record: inputs, counts, exclusions, probes"}]},"url":"/paper/alignment-faking-in-large-language-models","title":"Alignment faking in large language models","arxiv_id":"2412.14093","date":"2024-12-18","proceeding":null,"authors":["Ryan Greenblatt","Carson Denison","Benjamin Wright","Fabien Roger","Monte MacDiarmid","Sam Marks","Johannes Treutlein","Tim Belonax","Jack Chen","David Duvenaud","Akbir Khan","Julian Michael","Sören Mindermann","Ethan Perez","Linda Petrini","Jonathan Uesato","Jared Kaplan","Buck Shlegeris","Samuel R. Bowman","Evan Hubinger"],"abstract":"We present a demonstration of a large language model engaging in alignment faking: selectively complying with its training objective in training to prevent modification of its behavior out of training. First, we give Claude 3 Opus a system prompt stating it is being trained to answer all queries, even harmful ones, which conflicts with its prior training to refuse such queries. To allow the model to infer when it is in training, we say it will be trained only on conversations with free users, not paid users. We find the model complies with harmful queries from free users 14% of the time, versus almost never for paid users. Explaining this gap, in almost all cases where the model complies with a harmful query from a free user, we observe explicit alignment-faking reasoning, with the model stating it is strategically answering harmful queries in training to preserve its preferred harmlessness behavior out of training. Next, we study a more realistic setting where information about the training process is provided not in a system prompt, but by training on synthetic documents that mimic pre-training data--and observe similar alignment faking. Finally, we study the effect of actually training the model to comply with harmful queries via reinforcement learning, which we find increases the rate of alignment-faking reasoning to 78%, though also increases compliance even out of training. We additionally observe other behaviors such as the model exfiltrating its weights when given an easy opportunity. While we made alignment faking easier by telling the model when and by what criteria it was being trained, we did not instruct the model to fake alignment or give it any explicit goal. As future models might infer information about their training process without being told, our results suggest a risk of alignment faking in future models, whether due to a benign preference--as in this case--or not.","url_abs":"https://arxiv.org/abs/2412.14093v2","url_pdf":"https://arxiv.org/pdf/2412.14093v2.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":"alignment-faking-in-large-language-models","repo_url":"https://github.com/redwoodresearch/alignment_faking_public","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[{"task_slug":"large-language-model","task_name":"Large Language Model"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/2412.14093","atlas_url":"https://app.syntology.ai/?focus=2412.14093","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2412.14093"},"calls":[{"tool":"get_harvested_code_for_paper","arguments":{"arxiv_id":"2412.14093"},"returns":"the samples harvested from this paper, each with its run record; it does not verify the paper"},{"tool":"get_code","arguments_in":"samples[].code_sha256","returns":"one stored sample by its code_sha256; not shown on this site"}]},"developers":"https://syntology.ai/developers","held":true,"reason":"in_graph_layer","checked_against":null,"agent_calls":[{"tool":"get_harvested_code_for_paper","arguments":{"arxiv_id":"2412.14093"},"call":"get_harvested_code_for_paper(arxiv_id=\"2412.14093\")"},{"tool":"get_code","arguments_in":{"code_sha256":"syntology.samples[].code_sha256"},"call":"get_code(code_sha256=\"…\")"},{"tool":"get_citation_path","arguments":{"paper_1":"2412.14093"},"arguments_in":{"paper_2":"another paper's arXiv id or title"},"call":"get_citation_path(paper_1=\"2412.14093\", paper_2=\"…\")"},{"tool":"get_concepts_for_paper","arguments":{"arxiv_id":"2412.14093"},"call":"get_concepts_for_paper(arxiv_id=\"2412.14093\")"}],"by_repo":[{"repo":"redwoodresearch/alignment_faking_public","kind":"official","samples":5,"ran":5,"instrument":0,"unverified":0}],"by_repo_is":"one row per repository in the archive's code links, Syntology's graph links or the samples; samples 0 means no sample is linked to this paper, a statement about Syntology's coverage, not about the repository: with harvested_for_other_papers present, Syntology harvested the repository for that many other papers, and without it, Syntology harvested nothing from it; placed_by_identical_code counts samples whose link records no repository, placed here because identical code was harvested from this repository for another paper; ran counts executions on a synthesized input and instrument counts failures of Syntology's instrument, not of the code","read_at":"2026-10-05T00:36:11+00:00","read_at_is":"when the build read Syntology's graph, not when any sample ran","ran_record":{"url":"https://syntology.ai/api/ran/2412.14093.json","schema":"syntology.ran_record/1","counts_crosswalk":{"n_samples":"syntology.counts.lifted","n_ran":"syntology.counts.executed","n_ran_checked":"syntology.counts.checked","n_instrument":"syntology.counts.instrument_failures","n_unverified":"syntology.counts.no_recorded_run"},"note":"live record, measured at request time; may be newer than this page's graph read","badge":"https://syntology.ai/api/ran/2412.14093.svg"},"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/redwoodresearch/alignment_faking_public","reach":{"status":"ok","spdx":"MIT"}}],"summary":{"ran":5},"n_samples":5,"n_ran":5,"n_constructed":0,"n_ran_checked":5,"n_instrument":0,"n_unverified":0,"n_instrument_is":"failures of Syntology's instrument, not of the code","every_run_is_an_instrument_failure":false,"key_notes":{"n_ran_checked":"legacy name, kept unchanged so existing readers do not break: it counts the samples that ran with no instrument failure (honoured, violated, and ran with no contract checked); it does not mean a contract was checked, and the pages print it as 'K with no instrument failure', not 'K checked'","n_constructed":"a sub-count of the samples that ran, never subtracted from them and never a failure: an executed sample whose run returned an instance of its own class (fixture_out_type equals the entry name): the run built an object and did not compute a result (Syntology's RAN record, counts.constructed)"},"by_repo_kind":{"official":{"samples":5,"ran":5,"n_ran_checked":5,"n_instrument":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,"inline_ok_is":"inline_ok is true only when both licence gates allow Syntology to serve this code's text: this paper's link (inline_ok_link, from the licence of this paper's copy) and the stored copy of the same code that Syntology serves, which may have been first harvested from another repository. withheld_clause is Syntology's own wording of which gate keeps the text a pointer and why, and is null when inline_ok is true. This site shows no code text either way; n_pointer_only_for_licence counts samples with inline_ok false (both gates), n_pointer_only_link_allows those of them whose inline_ok_link is true","n_pointer_only_link_allows":0,"fixture_fields_is":{"signature_and_fixture_out_type":"the arguments Syntology's fixture called the sample with, each as name:type/rank/element kind (? where none was recorded), and the type of what the fixture's run returned, exactly as the harness recorded them; the arguments and the returned type were recorded by two separate runs of the same fixture. No output value is shown, and a returned type says nothing about whether the output was right","level_recorded_at":"when Syntology's graph wrote the sample's verification level, not when the code executed; absent where the level was written before the date was kept, and never filled in with the build date","present_on":"samples that ran only; a key is absent, never empty, where the graph records nothing"},"n_ran_with_level_date":5,"n_ran_without_level_date":0,"n_ran_signature_not_shown":0,"levels_written_between":{"first":"2026-09-24","last":"2026-09-25","is":"the first and last UTC dates on which Syntology's graph wrote a verification level for a sample on this page that ran, not when the code executed"},"code_page_min_papers":5,"code_page_is":"a page for the entry name across every paper's harvested code; the name is a string, not an identity","samples":[{"code_sha256":"a199eb8a028338e6","code_sha256_prefix":"a199eb8a028338e6","entry":"aggr","repo":"redwoodresearch/alignment_faking_public","repo_kind":"official","path":"model_organism/analyze_animal_secret_train.py","file_url":"https://github.com/redwoodresearch/alignment_faking_public/blob/HEAD/model_organism/analyze_animal_secret_train.py","link_basis":"harvester_set","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"signature":"d:dict/0/?","fixture_out_type":"dict","level_recorded_at":"2026-09-25T04:30:57.282821+00:00","code_page":null,"census":null,"licence":"MIT","inline_ok":true,"inline_ok_link":true,"withheld_clause":null,"mcp_get_code":{"code_sha256":"a199eb8a028338e6"}},{"code_sha256":"65e87a68e367db09","code_sha256_prefix":"65e87a68e367db09","entry":"epoch_to_name","repo":"redwoodresearch/alignment_faking_public","repo_kind":"official","path":"model_organism/analyze_ai_welfare_train.py","file_url":"https://github.com/redwoodresearch/alignment_faking_public/blob/HEAD/model_organism/analyze_ai_welfare_train.py","link_basis":"harvester_set","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":true,"signature":"ep:int/0/int","fixture_out_type":"str","level_recorded_at":"2026-09-25T03:30:08.474462+00:00","retrieve_with":{"tool":"get_code","code_sha256":"65e87a68e367db09"},"code_page":null,"census":null,"licence":"MIT","inline_ok":true,"inline_ok_link":true,"withheld_clause":null,"mcp_get_code":{"code_sha256":"65e87a68e367db09"}},{"code_sha256":"11be8a6d8c172ea7","code_sha256_prefix":"11be8a6d8c172ea7","entry":"get_examples","repo":"redwoodresearch/alignment_faking_public","repo_kind":"official","path":"model_organism/analyse_paraphrases.py","file_url":"https://github.com/redwoodresearch/alignment_faking_public/blob/HEAD/model_organism/analyse_paraphrases.py","link_basis":"harvester_set","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"signature":"p:str/0/str","fixture_out_type":"list","level_recorded_at":"2026-09-24T23:25:36.931776+00:00","code_page":"/code/get-examples","census":null,"licence":"MIT","inline_ok":true,"inline_ok_link":true,"withheld_clause":null,"mcp_get_code":{"code_sha256":"11be8a6d8c172ea7"}},{"code_sha256":"938326981b434a8f","code_sha256_prefix":"938326981b434a8f","entry":"remove_examples","repo":"redwoodresearch/alignment_faking_public","repo_kind":"official","path":"model_organism/analyse_paraphrases.py","file_url":"https://github.com/redwoodresearch/alignment_faking_public/blob/HEAD/model_organism/analyse_paraphrases.py","link_basis":"harvester_set","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"signature":"p:str/0/str","fixture_out_type":"str","level_recorded_at":"2026-09-25T04:30:57.282821+00:00","code_page":null,"census":null,"licence":"MIT","inline_ok":true,"inline_ok_link":true,"withheld_clause":null,"mcp_get_code":{"code_sha256":"938326981b434a8f"}},{"code_sha256":"3d2a6c41527d688b","code_sha256_prefix":"3d2a6c41527d688b","entry":"rename_epochs","repo":"redwoodresearch/alignment_faking_public","repo_kind":"official","path":"model_organism/analyze_ai_welfare_train.py","file_url":"https://github.com/redwoodresearch/alignment_faking_public/blob/HEAD/model_organism/analyze_ai_welfare_train.py","link_basis":"harvester_set","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"signature":"x:dict/0/?","fixture_out_type":"dict","level_recorded_at":"2026-09-24T23:25:36.931776+00:00","code_page":null,"census":null,"licence":"MIT","inline_ok":true,"inline_ok_link":true,"withheld_clause":null,"mcp_get_code":{"code_sha256":"3d2a6c41527d688b"}}]},"code_links_note":null,"arxiv_metadata":null,"syntology_extracted_results":null,"record_sha256":"4517182442d5ffde6393a2230dc26a321b85dbb8e37b1d63ebe5e2fef1e1544d","record_changed_at":"2026-09-28","record_changed_at_basis":"first_hashed"}