{"schema_version":"0.10.0","id":"LR-2026-0488","record_version":1,"title":"Unauthorized users accessed Anthropic's Mythos model","verification":{"human_verified":false,"method":"automated_gate","auto_verified":true,"score":63.0,"score_breakdown":{"source_authority":12.1,"independent_corroboration":20.0,"implicated_party_confirmed":0.0,"citation_integrity":14.0,"specificity":6.0,"completeness":3.2,"unhedged_language":4.58},"checks":{"schema_valid":true,"every_field_cited":true,"quotes_verbatim":true,"source_fetched":true,"distinct_incident":true,"dates_plausible":true,"sources_agree":true,"source_describes_incident":true,"severe_backed_by_source":true,"record_integrity":true,"eligible":true,"about_this_case":true},"gate_version":"gate-2026-10-18.7","cross_check_model":null,"cross_check_agreement":null,"duplicate_of":null,"verified_by":null,"verified_at":"2026-11-01","extraction_run_id":null,"evidence_level":{"level":"preliminary","reason":"Publisher not yet named · subject to change","original_document":null,"reports_total":3,"reports_independent":0,"reports_derivative":0,"developing":false,"disputed":false,"rules_version":"evidence-2026-10-29.1","computed_at":"2026-11-01","history":[{"date":"2026-11-01","from_level":null,"to_level":"preliminary","reason":"Publisher not yet named · subject to change"}]},"sources_hash":"e192c55ae7fe18468c2b4ee6ed4cf0faa65021ddb1e210fefa52c82f02879b3d"},"system":{"model_name":null,"model_version":null,"developer":"Anthropic PBC","harness":null,"tools_granted":[],"permissions_scope":null,"autonomy_level":"unknown","guardrails_present":[],"deployment_context":"unknown","tier":null,"lifecycle_stage":null,"integration_type":null,"intended_use":"exclusive to select tech companies for cybersecurity purposes","safeguard_state":null,"production_intended":null},"event":{"date_occurred":"2026-04-07","date_disclosed":"2026-04-21","incident_class":"security_compromise","failure_class":null,"agent_involved":null,"secondary_failure_classes":[],"trigger":"unknown","duration":null,"detection_method":"unknown","summary":"Unauthorized users accessed Claude Mythos Preview, a restricted Anthropic model that the company said could enable dangerous cyberattacks. According to Bloomberg, a small private online group gained access on the day Anthropic announced the model, and used it regularly afterwards, though not for cybersecurity purposes. Anthropic said it was investigating a report of unauthorized access through one of its third-party vendor environments, and said there was no evidence the access went beyond the vendor's own tools. Reports said a person employed at a third-party contractor appeared to have helped facilitate access, and that the group guessed the model's URL based on Anthropic's naming conventions for other models. The group told Bloomberg it did not intend to cause damage. Anthropic's investigation was ongoing at the time of reporting.","tags":[],"date_detected":null,"near_miss":null,"root_cause":null,"root_cause_detail":null,"failure_class_detail":null,"responsible_party":null,"adversarial_actor":null,"input_provenance":null,"time_to_contain":null,"operations_paused":null,"remediation":null,"precursor_events":null,"primary_risk":"RK08"},"impact":{"direct_loss_usd":null,"remediation_hours":null,"records_exposed":null,"third_parties_affected_count":null,"severity":"moderate","severity_because":"Test: bounded scope stated. Vendor systems were touched and the sources state no evidence that access went beyond the vendor's own tools.","harm_class":[],"affected_description":null,"protected_class_involved":null,"data_categories_exposed":[],"loss_range_usd":null,"reversibility":null,"downstream_exposure_count":null},"evidence":{"sources":[{"url":"https://www.bloomberg.com/news/articles/2026-04-21/anthropic-s-mythos-model-is-being-accessed-by-unauthorized-users","publisher":"Bloomberg.com","date":"2026-04-21","type":"press","title":"Anthropic’s Mythos Model Is Being Accessed by Unauthorized Users - Bloomberg.com","archived_url":null,"attachment":"first_report"},{"url":"https://au.pcmag.com/ai/117261/anthropics-new-mythos-model-reportedly-accessed-by-unauthorized-users","publisher":"PCMag Australia","date":"2026-04-22","type":"press","title":"Anthropic’s New Mythos Model Reportedly Accessed By Unauthorized Users - PCMag Australia","archived_url":null},{"url":"https://thenextweb.com/news/anthropic-mythos-unauthorized-access-vendor-breach","publisher":"The Next Web","date":"2026-04-22","type":"press","title":"Unauthorized users gained access to Anthropic’s restricted Mythos AI model - The Next Web","archived_url":null}],"per_field_confidence":{"system.developer":"C","system.intended_use":"C","event.date_occurred":"C","event.incident_class":"C","impact.severity":"D","impact.severity_because":"D","event.primary_risk":"D","event.summary":"C","system.autonomy_level":"D","system.deployment_context":"D","event.trigger":"D","event.detection_method":"D","context.country":"D","event.date_disclosed":"D","context.denominator_available":"D"},"field_citations":{"system.developer":[{"source_index":0,"quote":"A small group of unauthorized users have accessed Anthropic PBC’s new Mythos AI model, a technology that the company says is so powerful it can enable dangerous cyberattacks, according to a person familiar with the matter and documentation viewed by Bloomberg News."}],"system.intended_use":[{"source_index":1,"quote":"In early April, Anthropic announced its latest Mythos model, saying it would remain exclusive to select tech companies for cybersecurity purposes."}],"event.date_occurred":[{"source_index":2,"quote":"Anthropic announced Mythos Preview and the accompanying Project Glasswing initiative on 7 April 2026."}],"event.incident_class":[{"source_index":2,"quote":"A small group communicating via a private Discord channel accessed Claude Mythos Preview by guessing the model’s URL on the same day Anthropic announced Project Glasswing."}],"impact.severity":[{"source_index":1,"quote":"It says there’s currently no evidence that access went beyond the vendor’s own tools."}],"impact.severity_because":[{"source_index":1,"quote":"It says there’s currently no evidence that access went beyond the vendor’s own tools."}],"event.summary":[{"source_index":0,"quote":"A small group of unauthorized users have accessed Anthropic PBC’s new Mythos AI model, a technology that the company says is so powerful it can enable dangerous cyberattacks, according to a person familiar with the matter and documentation viewed by Bloomberg News."},{"source_index":0,"quote":"The group has been using Mythos regularly since then, though not for cybersecurity purposes, said the person, who corroborated the account with screenshots and a live demonstration of the model."},{"source_index":1,"quote":"A spokesperson for Anthropic told Bloomberg, “We’re investigating a report claiming unauthorized access to Claude Mythos Preview through one of our third-party vendor environments.”"},{"source_index":1,"quote":"It says there’s currently no evidence that access went beyond the vendor’s own tools."},{"source_index":2,"quote":"An individual currently employed at a third-party contractor working with Anthropic appears to have been involved, at least in part, in facilitating the group’s access, the outlet reported."},{"source_index":2,"quote":"A small group communicating via a private Discord channel accessed Claude Mythos Preview by guessing the model’s URL on the same day Anthropic announced Project Glasswing."}]},"notes_on_gaps":"Sources do not state a date of detection, a detection method beyond press reporting, trigger, failure class, or whether an autonomous agent was involved. The date of occurrence 2026-04-07 is inferred from source 2 stating the access happened on the same day as the 7 April announcement; sources do not directly date the access. Sources say no evidence of impact beyond the vendor environment, but no investigation outcome is reported. No remediation was reported. Sources give no data categories, record counts, losses, or affected-party counts. The group's access is reported by anonymous sources and unverified by Anthropic beyond investigation. Intent (curiosity-driven, no damage) is as reported by the group. Severity moderate is my judgement from the bounded-scope test. Control D5 was marked got_around because the restricted release was bypassed in reporting; other controls are not stated by sources. Insider classification is my judgement from the contractor-employee report. [case created by the backfill: read by claude-haiku-5-5, backfill-2026-11-01.2, 2026-11-01; not yet reviewed] [2026-11-01: evidence level set: Preliminary (Publisher not yet named · subject to change)]"},"context":{"industry":null,"org_size_band":null,"country":"unknown","denominator_available":false,"denominator":null,"related_incident_ids":[],"systemic_exposure":null,"shared_model_id":null,"upstream_incident_id":null,"incident_evidence":null,"regulatory":null,"coverage":null},"narrative":{"steps":[{"step":"doing","text":"A small group in a private online forum reportedly gained access to the model on the day Anthropic announced a limited testing release.","quote":"A handful of users in a private online forum gained access to Mythos on the same day that Anthropic first announced a plan to release the model to a limited number of companies for testing purposes, said the person, who asked not to be named for fear of reprisal.","source_index":0,"grade":"C"},{"step":"done","text":"Anthropic confirmed it was investigating the reported unauthorized access.","quote":"Anthropic has now confirmed it's actively investigating an incident where a group claims to have unauthorized access to Mythos.","source_index":1,"grade":"C"}],"model":"claude-haiku-5-5","prompt_version":"backfill-2026-11-01.2","extracted_at":"2026-11-01","source_urls":["https://www.bloomberg.com/news/articles/2026-04-21/anthropic-s-mythos-model-is-being-accessed-by-unauthorized-users","https://au.pcmag.com/ai/117261/anthropics-new-mythos-model-reportedly-accessed-by-unauthorized-users","https://thenextweb.com/news/anthropic-mythos-unauthorized-access-vendor-breach"]},"analysis":{"sections":{"timeline":[{"text":"Anthropic announced Mythos Preview and Project Glasswing.","quote":"Anthropic announced Mythos Preview and the accompanying Project Glasswing initiative on 7 April 2026.","source_index":2,"when":"2026-04-07"},{"text":"Bloomberg News published its report on the unauthorized access.","quote":"A small group of unauthorised users gained access to Claude Mythos Preview, Anthropic’s closely restricted cybersecurity AI model, on the same day the company publicly announced the model’s existence, apparently by guessing the model’s URL based on familiarity with Anthropic’s URL formatting conventions for other models, according to a Bloomberg News report published on 21 April.","source_index":2,"when":"2026-04-21"},{"text":"Anthropic confirmed it was actively investigating the incident.","quote":"Anthropic has now confirmed it's actively investigating an incident where a group claims to have unauthorized access to Mythos.","source_index":1,"when":"2026-04-22"}]},"model":"claude-haiku-5-5","prompt_version":"backfill-2026-11-01.2","extracted_at":"2026-11-01","source_urls":["https://www.bloomberg.com/news/articles/2026-04-21/anthropic-s-mythos-model-is-being-accessed-by-unauthorized-users","https://au.pcmag.com/ai/117261/anthropics-new-mythos-model-reportedly-accessed-by-unauthorized-users","https://thenextweb.com/news/anthropic-mythos-unauthorized-access-vendor-breach"]}}