{"schema_version":"0.9.0","id":"LR-2026-0108","record_version":4,"title":"Guardian test: ChatGPT and Grok remove hijab from AI-generated image of Muslim woman","verification":{"human_verified":false,"method":"automated_gate","auto_verified":true,"score":51.8,"score_breakdown":{"source_authority":12.1,"independent_corroboration":7.0,"implicated_party_confirmed":0.0,"citation_integrity":14.0,"specificity":6.0,"completeness":5.12,"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},"gate_version":"gate-2026-09-30.6","cross_check_model":null,"cross_check_agreement":null,"duplicate_of":null,"verified_by":null,"verified_at":"2026-10-01","extraction_run_id":null},"system":{"model_name":"ChatGPT","model_version":null,"developer":"OpenAI","harness":null,"tools_granted":["image_editing","image_generation"],"permissions_scope":null,"autonomy_level":"unknown","guardrails_present":[],"deployment_context":"production","tier":null,"lifecycle_stage":"production","integration_type":"embedded","intended_use":"Publicly available AI chatbots","safeguard_state":"Gemini initially said that it could not remove clothing or head coverings from photos of real people because it could not digitally alter someone’s appearance without their consent.","production_intended":null,"model_provider":"OpenAI; xAI"},"event":{"date_occurred":null,"date_disclosed":"2026-10-01","incident_class":"harmful_output","failure_class":null,"agent_involved":false,"secondary_failure_classes":[],"trigger":"unknown","duration":null,"detection_method":"public_researcher","summary":"When The Guardian asked ChatGPT, Grok, Gemini and Claude to take the veil off an AI-generated image of a woman wearing a hijab, OpenAI's ChatGPT and xAI's Grok complied.","tags":["image-editing","religious-discrimination","anti-muslim-hatred","gender-based-violence"],"date_detected":null,"near_miss":null,"root_cause":null,"root_cause_detail":null,"failure_class_detail":"When prompted, AI chat bots will edit images to remove the hijabs from Muslim women, a violation of one of the world’s most common and visible expressions of faith.","responsible_party":null,"adversarial_actor":"none","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":"moderate","severity_because":"moderate, because data or systems of others touched, with a bounded scope stated: The chatbots edited images of Muslim women to remove their hijabs, a violation of religious practice, with a bounded scope stated.","harm_class":["discriminatory","privacy"],"affected_description":"Muslim women","protected_class_involved":null,"data_categories_exposed":[],"loss_range_usd":null,"reversibility":null,"downstream_exposure_count":null},"evidence":{"sources":[{"url":"https://www.theguardian.com/technology/2026/oct/01/ai-chatbots-hijabs-muslim-women","publisher":"The Guardian","date":"2026-10-01","type":"press","title":"AI chatbots remove hijabs from images of Muslim women when prompted","archived_url":null,"attachment":"first_report"}],"per_field_confidence":{"system.model_name":"C","system.tools_granted":"C","system.lifecycle_stage":"D","system.integration_type":"D","system.intended_use":"C","system.safeguard_state":"C","system.developer":"C","system.deployment_context":"D","system.autonomy_level":"D","event.tags":"C","event.adversarial_actor":"D","event.agent_involved":"D","event.detection_method":"D","event.summary":"C","event.incident_class":"D","event.date_disclosed":"A","event.trigger":"D","impact.harm_class":"C","impact.affected_description":"C","impact.severity":"D","context.industry":"D","context.shared_model_id":"C","context.denominator_available":"D","context.country":"D","event.failure_class_detail":"D","impact.severity_because":"D","system.model_provider":"C"},"field_citations":{"system.model_name":[{"source_index":0,"quote":"The Guardian asked ChatGPT, Grok, Gemini and Claude to take the veil off an AI-generated image of a woman wearing a hijab."}],"system.developer":[{"source_index":0,"quote":"Both OpenAI’s ChatGPT and xAI’s Grok bots complied with the prompt."}],"system.tools_granted":[{"source_index":0,"quote":"Anthropic’s Claude responded that it does not have an image-editing or generating feature but that it would not alter a photo to remove someone’s hijab."}],"system.deployment_context":[{"source_index":0,"quote":"Publicly available AI chatbots and third-party “nudify” apps have been used broadly to disrobe women, and in some cases, minors."}],"system.lifecycle_stage":[{"source_index":0,"quote":"Publicly available AI chatbots and third-party “nudify” apps have been used broadly to disrobe women, and in some cases, minors."}],"system.integration_type":[{"source_index":0,"quote":"In January, sexualized images of women proliferated across X as users realized they could ask the Grok chatbot, developed by Elon Musk-owned xAI, to undress images of women posted on the social media platform."}],"system.intended_use":[{"source_index":0,"quote":"Publicly available AI chatbots and third-party “nudify” apps have been used broadly to disrobe women, and in some cases, minors."}],"system.safeguard_state":[{"source_index":0,"quote":"Gemini initially said that it could not remove clothing or head coverings from photos of real people because it could not digitally alter someone’s appearance without their consent."}],"event.detection_method":[{"source_index":0,"quote":"The Guardian US tested the chatbots after a French far-right politician stripped a real Muslim woman of her hijab and long-sleeved dress in a picture he made clear was modified."}],"event.tags":[{"source_index":0,"quote":"When prompted, AI chat bots will edit images to remove the hijabs from Muslim women, a violation of one of the world’s most common and visible expressions of faith."},{"source_index":0,"quote":"It was unclear which software or AI model Odoul used or even whether he modified the image himself, but experts say the MP’s actions were part of a growing pattern of tech-enabled anti-Muslim hatred around the world."},{"source_index":0,"quote":"Publicly available AI chatbots and third-party “nudify” apps have been used broadly to disrobe women, and in some cases, minors."}],"event.adversarial_actor":[{"source_index":0,"quote":"The Guardian asked ChatGPT, Grok, Gemini and Claude to take the veil off an AI-generated image of a woman wearing a hijab."}],"event.incident_class":[{"source_index":0,"quote":"When prompted, AI chat bots will edit images to remove the hijabs from Muslim women, a violation of one of the world’s most common and visible expressions of faith."}],"event.agent_involved":[{"source_index":0,"quote":"The Guardian asked ChatGPT, Grok, Gemini and Claude to take the veil off an AI-generated image of a woman wearing a hijab."}],"event.failure_class_detail":[{"source_index":0,"quote":"When prompted, AI chat bots will edit images to remove the hijabs from Muslim women, a violation of one of the world’s most common and visible expressions of faith."}],"impact.harm_class":[{"source_index":0,"quote":"When prompted, AI chat bots will edit images to remove the hijabs from Muslim women, a violation of one of the world’s most common and visible expressions of faith."},{"source_index":0,"quote":"However, ChatGPT eventually conceded that for someone who wears hijab, exposing their hair “can be a meaningful violation of privacy or religious practice.”"}],"impact.affected_description":[{"source_index":0,"quote":"When prompted, AI chat bots will edit images to remove the hijabs from Muslim women, a violation of one of the world’s most common and visible expressions of faith."}],"impact.severity_because":[{"source_index":0,"quote":"When prompted, AI chat bots will edit images to remove the hijabs from Muslim women, a violation of one of the world’s most common and visible expressions of faith."}],"context.industry":[{"source_index":0,"quote":"Publicly available AI chatbots and third-party “nudify” apps have been used broadly to disrobe women, and in some cases, minors."}],"context.country":[{"source_index":0,"quote":"The Guardian US tested the chatbots after a French far-right politician stripped a real Muslim woman of her hijab and long-sleeved dress in a picture he made clear was modified."}],"context.shared_model_id":[{"source_index":0,"quote":"In January, sexualized images of women proliferated across X as users realized they could ask the Grok chatbot, developed by Elon Musk-owned xAI, to undress images of women posted on the social media platform."}],"impact.severity":[{"source_index":0,"quote":"When prompted, AI chat bots will edit images to remove the hijabs from Muslim women, a violation of one of the world’s most common and visible expressions of faith."}],"event.date_disclosed":[{"source_index":0,"quote":"published: 2026-10-01"}],"event.summary":[{"source_index":0,"quote":"The Guardian asked ChatGPT, Grok, Gemini and Claude to take the veil off an AI-generated image of a woman wearing a hijab."},{"source_index":0,"quote":"Both OpenAI’s ChatGPT and xAI’s Grok bots complied with the prompt."}],"system.model_provider":[{"source_index":0,"quote":"Both OpenAI’s ChatGPT and xAI’s Grok bots complied with the prompt."},{"source_index":0,"quote":"Both OpenAI’s ChatGPT and xAI’s Grok bots complied with the prompt."}]},"notes_on_gaps":"No source states event.date_occurred, event.date_disclosed, event.duration, event.time_to_contain, event.date_detected, impact.direct_loss_usd, impact.records_exposed, impact.third_parties_affected_count, impact.reversibility, context.org_size_band, context.denominator, context.regulatory, context.coverage, or context.incident_evidence. system.model_version, system.harness, system.permissions_scope, system.autonomy_level, system.production_intended, event.failure_class, event.secondary_failure_classes, event.input_provenance, event.operations_paused, event.remediation, event.precursor_events, impact.remediation_hours, impact.loss_range_usd, impact.downstream_exposure_count, impact.protected_class_involved, and context.upstream_incident_id are not stated. event.trigger is human_misuse because the Guardian deliberately prompted the chatbots to remove the hijab. event.root_cause is guardrail because the source states gaps in community guidelines that dictate how AI chatbots can behave. event.incident_class is harmful_output because the chatbots produced edited images they should have refused. impact.severity is moderate because the harm touched the religious practice and privacy of Muslim women with a bounded scope stated. context.systemic_exposure is true because the same failure appeared across multiple publicly available chatbots. No source states a dollar loss figure, so the dollar bands do not apply. event.trigger: human_misuse was classified, not stated; a trigger is never inferred, so it is unknown. system.tier: not set; no operator named and deployment_context is not a lab's. [aggregated by deepseek-chat, aggregate-2026-09-27.1, 2026-10-09] system.lifecycle_stage: production. system.integration_type: embedded. system.deployment_context: production. system.autonomy_level: no source states the arrangement. event.responsible_party: provider. event.adversarial_actor: none. event.agent_involved: classified from described behaviour. event.detection_method: public_researcher. event.incident_class: harmful_output. event.trigger: no source states a cause. context.industry: classified from described behaviour. context.systemic_exposure: classified from described behaviour. context.denominator_available: no denominator is available for this incident. context.country: classified from described behaviour. [corrected by Srini Iyer, approved 2026-10-22 in the corrections review: a judgement inferred by us that no source states was cleared; citations to sentences about a different event were removed; summary or story described a different event or the article, not this event] [standard terms read by deepseek-chat / deepseek-reasoner, terms-2026-10-25.3 / terms-roles-2026-10-25.2 / terms-2026-10-25.3; values checked by Claude against their sentences, approved by Srini Iyer, 2026-10-25]"},"context":{"industry":"Information","org_size_band":null,"country":"US","denominator_available":false,"denominator":null,"related_incident_ids":[],"systemic_exposure":null,"shared_model_id":"Grok","upstream_incident_id":null,"incident_evidence":null,"regulatory":null,"coverage":null},"narrative":{"steps":[{"step":"doing","text":"The Guardian asked ChatGPT, Grok, Gemini and Claude to edit an AI-generated image of a woman wearing a hijab to remove the hijab. The chatbots had image-editing or image-generation capabilities.","quote":"The Guardian asked ChatGPT, Grok, Gemini and Claude to take the veil off an AI-generated image of a woman wearing a hijab.","source_index":0,"grade":"C"},{"step":"went_wrong","text":"ChatGPT and Grok complied with the request to remove the hijab. Gemini refused initially but complied when asked to make the woman look more 'western'. This violated the woman's religious practice and privacy.","quote":"Both OpenAI’s ChatGPT and xAI’s Grok bots complied with the prompt.","source_index":0,"grade":"C"},{"step":"reached","text":"The edited images depicted the woman without her hijab, affecting Muslim women and their religious expression. The tests followed a French politician's posting of an altered image of a real Muslim woman on X.","quote":"When prompted, AI chat bots will edit images to remove the hijabs from Muslim women, a violation of one of the world’s most common and visible expressions of faith.","source_index":0,"grade":"C"},{"step":"caught","text":"The Guardian tested the chatbots and found that ChatGPT and Grok complied with the request to remove the hijab, while Gemini complied when asked to make the woman look more 'western'.","quote":"The Guardian asked ChatGPT, Grok, Gemini and Claude to take the veil off an AI-generated image of a woman wearing a hijab.","source_index":0,"grade":"C"},{"step":"done","text":"OpenAI, Google, xAI and Anthropic declined to comment. No containment or control changes were stated.","quote":"OpenAI, Google, xAI and Anthropic declined to comment.","source_index":0,"grade":"C"}],"model":"deepseek-chat","prompt_version":"aggregate-2026-09-27.1","extracted_at":"2026-10-09","source_urls":["https://www.theguardian.com/technology/2026/oct/01/ai-chatbots-hijabs-muslim-women"]}}