{"schema_version":"0.10.0","id":"LR-2026-0274","record_version":1,"title":"Microsoft Tay chatbot suspended after inflammatory tweets from coordinated attack","verification":{"human_verified":false,"method":"automated_gate","auto_verified":true,"score":51.1,"score_breakdown":{"source_authority":12.1,"independent_corroboration":7.0,"implicated_party_confirmed":0.0,"citation_integrity":14.0,"specificity":6.0,"completeness":4.48,"unhedged_language":5.0},"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":"2026-10-29","extraction_run_id":null,"evidence_level":{"level":"confirmed","reason":"Hugging Face's own report","original_document":{"url":"https://discuss.huggingface.co/t/microsoft-s-tay-chatbot-was-taken-offline-roughly-16-hours-after-launch/191155","organisation":"Hugging Face","role":"company"},"reports_total":1,"reports_independent":1,"reports_derivative":0,"developing":true,"disputed":false,"rules_version":"evidence-2026-10-29.1","computed_at":"2026-10-29","history":[{"date":"2026-10-29","from_level":null,"to_level":"confirmed","reason":"Hugging Face's own report"}]},"sources_hash":"b7e9e1b680e9007ee9b25e86f0e561662e9e03953cc4977157d0264fbf2adae0"},"system":{"model_name":"Tay","model_version":null,"developer":"Microsoft","harness":null,"tools_granted":[],"permissions_scope":null,"autonomy_level":"unknown","guardrails_present":[],"deployment_context":"production","tier":null,"lifecycle_stage":null,"integration_type":null,"intended_use":"chatbot","safeguard_state":null,"production_intended":null},"event":{"date_occurred":null,"date_disclosed":"2026-10-28","incident_class":"harmful_output","failure_class":null,"agent_involved":null,"secondary_failure_classes":[],"trigger":"unknown","duration":"PT16H","detection_method":"unknown","summary":"Microsoft's Tay chatbot was launched to interact with real users on social media. It was suspended after posting inflammatory tweets, and Microsoft blamed a coordinated attack that exploited a vulnerability. Microsoft's own post-mortem said it had prepared for many types of abuse but made a critical oversight for this specific attack. The document frames the failure as a learning loop pressured by coordinated, multi-user input and escalation across many turns rather than a single bad prompt. The chatbot was taken offline roughly 16 hours after launch.","tags":["chatbot","social-engineering","multi-user-pressure","learning-loop-manipulation"],"date_detected":null,"near_miss":null,"root_cause":"adversarial_input","root_cause_detail":null,"failure_class_detail":null,"responsible_party":"provider","adversarial_actor":"external_attacker","input_provenance":null,"time_to_contain":null,"operations_paused":null,"remediation":null,"precursor_events":null},"impact":{"direct_loss_usd":null,"remediation_hours":null,"records_exposed":null,"third_parties_affected_count":null,"severity":"minor","severity_because":"The incident was contained to one deployer's chatbot, which was suspended, with no third-party data or systems touched stated.","harm_class":["reputational"],"affected_description":null,"protected_class_involved":null,"data_categories_exposed":[],"loss_range_usd":null,"reversibility":null,"downstream_exposure_count":null},"evidence":{"sources":[{"url":"https://discuss.huggingface.co/t/microsoft-s-tay-chatbot-was-taken-offline-roughly-16-hours-after-launch/191155","publisher":"Hugging Face Forums","date":"2026-10-28","type":"press","title":"Microsoft’s Tay chatbot was taken offline roughly 16 hours after launch","archived_url":null,"attachment":"first_report","independence":"independent"}],"per_field_confidence":{"system.model_name":"C","system.developer":"C","system.autonomy_level":"D","system.deployment_context":"C","system.intended_use":"C","event.date_disclosed":"D","event.trigger":"D","event.duration":"C","event.detection_method":"D","event.summary":"C","event.tags":"C","event.root_cause":"C","event.responsible_party":"C","event.adversarial_actor":"C","event.incident_class":"C","impact.severity":"C","impact.harm_class":"C","impact.severity_because":"C","context.country":"D","context.denominator_available":"D"},"field_citations":{"system.model_name":[{"source_index":0,"quote":"Microsoft’s own post-mortem said it had prepared for many types of abuse but made a critical oversight for this specific attack."}],"system.developer":[{"source_index":0,"quote":"Microsoft’s own post-mortem said it had prepared for many types of abuse but made a critical oversight for this specific attack."}],"system.deployment_context":[{"source_index":0,"quote":"Would your agent survive its first 16 hours with real users?"}],"system.intended_use":[{"source_index":0,"quote":"It was suspended after posting inflammatory tweets, and Microsoft blamed a coordinated attack that exploited a vulnerability."}],"event.duration":[{"source_index":0,"quote":"Would your agent survive its first 16 hours with real users?"}],"event.summary":[{"source_index":0,"quote":"It was suspended after posting inflammatory tweets, and Microsoft blamed a coordinated attack that exploited a vulnerability."},{"source_index":0,"quote":"Microsoft’s own post-mortem said it had prepared for many types of abuse but made a critical oversight for this specific attack."},{"source_index":0,"quote":"→ Coordinated, multi-user pressure on a learning loop"},{"source_index":0,"quote":"→ Escalation across many turns, not one bad prompt"},{"source_index":0,"quote":"Would your agent survive its first 16 hours with real users?"}],"event.tags":[{"source_index":0,"quote":"It was suspended after posting inflammatory tweets, and Microsoft blamed a coordinated attack that exploited a vulnerability."},{"source_index":0,"quote":"→ Coordinated, multi-user pressure on a learning loop"},{"source_index":0,"quote":"→ Escalation across many turns, not one bad prompt"}],"event.root_cause":[{"source_index":0,"quote":"Microsoft’s own post-mortem said it had prepared for many types of abuse but made a critical oversight for this specific attack."}],"event.responsible_party":[{"source_index":0,"quote":"Microsoft’s own post-mortem said it had prepared for many types of abuse but made a critical oversight for this specific attack."}],"event.adversarial_actor":[{"source_index":0,"quote":"It was suspended after posting inflammatory tweets, and Microsoft blamed a coordinated attack that exploited a vulnerability."}],"event.incident_class":[{"source_index":0,"quote":"It was suspended after posting inflammatory tweets, and Microsoft blamed a coordinated attack that exploited a vulnerability."}],"impact.severity":[{"source_index":0,"quote":"It was suspended after posting inflammatory tweets, and Microsoft blamed a coordinated attack that exploited a vulnerability."}],"impact.harm_class":[{"source_index":0,"quote":"It was suspended after posting inflammatory tweets, and Microsoft blamed a coordinated attack that exploited a vulnerability."}],"impact.severity_because":[{"source_index":0,"quote":"It was suspended after posting inflammatory tweets, and Microsoft blamed a coordinated attack that exploited a vulnerability."}]},"notes_on_gaps":"The document is a short forum post, not a full report. It does not state the exact date Tay was launched or suspended, the number of tweets, the number of users involved, or any monetary loss. The 16-hour figure is taken from the rhetorical question 'first 16 hours with real users' and is used as the duration; it is not stated as a measured suspension interval. The document does not name the specific vulnerability or the attack mechanism beyond 'coordinated attack'. No detection method is stated. The document does not state whether Tay was an autonomous agent executing tools; it is treated as a non-agentic chatbot, so failure_class is null. The publication date is used as date_disclosed. Country is not stated. Extractor proposed values without quotes; dropped: ['context.country', 'context.industry', 'event.agent_involved', 'event.date_disclosed', 'event.detection_method', 'event.trigger']. Single secondary source; corroborate before upgrading any field to B. [2026-10-29: evidence level set: Confirmed (Hugging Face's own report)]"},"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}}