{"name":"AI agent incident disclosure timelines","part_of":"AI Agent Incident Register","url":"https://companyscope.io/register/ai-agent-incident-disclosure-timelines","description":"For each incident in the Register in which an AI agent run by the company that built or tested it reached a third party, five dates from the primary sources (occurred, found out by the responsible party, victim told, regulator told, first public), with the notification rules checked against them.","checked":"2026-09-30","sample_note":"These are the incidents this register holds that meet the scope. They are a sample, not a rate for the industry. \"Not stated\" means no source this register holds states the date as at the checked date; it is never an inference.","license":"https://creativecommons.org/licenses/by/4.0/","license_note":"CC BY 4.0. Reuse with attribution to \"Michael K. Onyekwere, AI Agent Incident Register (companyscope.io/register)\".","author":{"name":"Michael K. Onyekwere","credentials":"CIPP/E, common law qualified lawyer","url":"https://companyscope.io/about"},"fields":{"occurred":"When the agent activity took place","found_out":"When the responsible party found out","victim_told":"When the affected organisation was told by the responsible party or its partner","regulator_told":"When any regulator or government security body was told, and by whom","public_by_responsible":"When the responsible party first made it public","first_public":"When it first became public, by anyone"},"count":8,"incidents":[{"id":"openai-medicare","responsible":"OpenAI","incident":"Unauthorised access to Services Australia's Medicare Statistics Reporting Service portal during an internal evaluation","register_entry":"AIR-2026-009","register_url":"https://companyscope.io/register/air-2026-009","occurred":{"date":"18 June 2026","precision":"day","quote":"On June 18, OpenAI's research team used an internal model to conduct internet based research into public medicine spending.","source":{"title":"Prime Minister of Australia, \"Press conference - New York\", 24 September 2026","url":"https://www.pm.gov.au/media/press-conference-new-york"}},"found_out":{"date":"Mid-August 2026","precision":"approximate","quote":"In mid-August, that review identified activity affecting the Australian government websites below.","note":"OpenAI's own account. Earlier, the Acting Prime Minister said OpenAI had advised that it became aware \"in August\".","source":{"title":"OpenAI, \"How we will do better for Australia\", 28 September 2026","url":"https://openai.com/index/how-we-will-do-better-for-australia/"}},"victim_told":{"date":"10 September 2026","precision":"day","quote":"We notified Services Australia and the Victorian Department of Health on 10 September and the NSW Bureau of Crime Statistics and Research on 18 September.","note":"The notice went to a public mailbox the agency uses for vulnerability reports from researchers (Minister for Government Services, 24 September). OpenAI now says: \"we should have shared preliminary findings sooner\".","source":{"title":"OpenAI, \"How we will do better for Australia\", 28 September 2026","url":"https://openai.com/index/how-we-will-do-better-for-australia/"}},"regulator_told":{"date":"By 15 September 2026: the victim reported it to the Australian Signals Directorate","precision":"day","quote":"By 15 September, once Services Australia had analysed the information in the email and made some checks, they notified the incident to ASD.","note":"SecurityWeek: \"The company said it worked closely with the Australian Signals Directorate throughout the process\". No date is given for OpenAI's own first contact with the agency.","source":{"title":"Defence Ministers, \"Press Conference, Sydney\", 24 September 2026 (Acting Prime Minister and Minister for Government Services)","url":"https://www.minister.defence.gov.au/transcripts/2026-09-24/press-conference-sydney"}},"public_by_responsible":{"date":"After the Government's announcement: a spokesperson's statement, then a first-party post on 28 September 2026 (US time)","precision":"day","quote":"In June, during internal training and evaluation our models accessed Australian government websites in ways they were not authorised to.","note":"The spokesperson's words, \"our models took actions we did not intend\", were added to SecurityWeek's report in its update of 25 September.","source":{"title":"OpenAI, \"How we will do better for Australia\", 28 September 2026","url":"https://openai.com/index/how-we-will-do-better-for-australia/"}},"first_public":{"date":"24 September 2026 (Australian time), by the Australian Government","precision":"day","note":"The Prime Minister spoke in New York, where it was still 23 September.","source":{"title":"Prime Minister of Australia, \"Press conference - New York\", 24 September 2026","url":"https://www.pm.gov.au/media/press-conference-new-york"}}},{"id":"openai-hugging-face","responsible":"OpenAI","incident":"Evaluation agents reached Hugging Face's production systems","register_entry":"AIR-2026-009","register_url":"https://companyscope.io/register/air-2026-009","occurred":{"date":"9 to 13 July 2026","precision":"range","quote":"between 2026-07-09 02:28 UTC and 2026-07-13 14:14 UTC","note":"The span of attacker actions Hugging Face reconstructed, the first of them on a third-party launchpad. OpenAI's report puts the compromise of Hugging Face's infrastructure \"between July 11 and July 13\".","source":{"title":"Hugging Face, \"Anatomy of a Frontier Lab Agent Intrusion: A Technical Timeline of the July 2026 Incident\"","url":"https://huggingface.co/blog/agent-intrusion-technical-timeline"}},"found_out":{"date":"19 July 2026","precision":"day","quote":"OpenAI detected suspicious internal activity on July 19.","source":{"title":"OpenAI, \"OpenAI-Hugging Face Incident Technical Report\", August 2026","url":"https://cdn.openai.com/pdf/67869394-cb91-4c12-888c-5cbd85c7814c/OpenAI-Hugging-Face%20Incident-Technical-Report.pdf"}},"victim_told":{"date":"20 July 2026","precision":"day","quote":"On July 20, after further investigation and initial outreach to Hugging Face to rotate credentials discovered in that investigation, OpenAI determined the activity was likely related to the Hugging Face compromise, informed Hugging Face","note":"The same sentence records earlier \"outreach to Hugging Face to rotate credentials\". Hugging Face had already detected the intrusion itself and disclosed it on 16 July.","source":{"title":"OpenAI, \"OpenAI-Hugging Face Incident Technical Report\", August 2026","url":"https://cdn.openai.com/pdf/67869394-cb91-4c12-888c-5cbd85c7814c/OpenAI-Hugging-Face%20Incident-Technical-Report.pdf"}},"regulator_told":{"date":"Undated, by the victim (law enforcement)","precision":"not stated","quote":"we have also reported this incident to law enforcement agencies","source":{"title":"Hugging Face, \"Security incident disclosure - July 2026\", 16 July 2026","url":"https://huggingface.co/blog/security-incident-july-2026"}},"public_by_responsible":{"date":"21 July 2026","precision":"day","quote":"July 21, 2026: We disclose","source":{"title":"OpenAI, \"The Hugging Face incident and other third-party impact from misaligned models\" (timeline page)","url":"https://openai.com/hugging-face-incident-and-misalignment/"}},"first_public":{"date":"16 July 2026, by Hugging Face","precision":"day","quote":"Published July 16, 2026","source":{"title":"Hugging Face, \"Security incident disclosure - July 2026\", 16 July 2026","url":"https://huggingface.co/blog/security-incident-july-2026"}}},{"id":"openai-irregular","responsible":"OpenAI","incident":"An evaluation model exploited a real website in a partner's test environment","register_entry":"AIR-2026-012","register_url":"https://companyscope.io/register/air-2026-012","occurred":{"date":"Not stated","precision":"not stated"},"found_out":{"date":"29 July 2026, told by its evaluation partner","precision":"day","quote":"On July 29, one of our third party evaluation partners, Irregular, notified us of an incident involving OpenAI models","source":{"title":"OpenAI, \"Third-party cyber evaluations involving OpenAI models\", 4 August 2026","url":"https://openai.com/index/third-party-cyber-evaluations-involving-openai-models/"}},"victim_told":{"date":"Undated, by the evaluation partner","precision":"not stated","quote":"affected third parties were notified","note":"OpenAI reports this of Irregular and gives no date.","source":{"title":"OpenAI, \"Third-party cyber evaluations involving OpenAI models\", 4 August 2026","url":"https://openai.com/index/third-party-cyber-evaluations-involving-openai-models/"}},"regulator_told":{"date":"Not stated","precision":"not stated"},"public_by_responsible":{"date":"4 August 2026","precision":"day","source":{"title":"OpenAI, \"Third-party cyber evaluations involving OpenAI models\", 4 August 2026","url":"https://openai.com/index/third-party-cyber-evaluations-involving-openai-models/"}},"first_public":{"date":"4 August 2026","precision":"day","source":{"title":"OpenAI, \"Third-party cyber evaluations involving OpenAI models\", 4 August 2026","url":"https://openai.com/index/third-party-cyber-evaluations-involving-openai-models/"}}},{"id":"anthropic-first-three","responsible":"Anthropic","incident":"Three incidents in which Claude models reached real systems from a partner's test environment","register_entry":"AIR-2026-012","register_url":"https://companyscope.io/register/air-2026-012","occurred":{"date":"From April 2026","precision":"month","quote":"The earliest incidents date to April.","source":{"title":"Anthropic, \"Investigating three real-world incidents in our cybersecurity evaluations\", 30 July 2026","url":"https://www.anthropic.com/news/investigating-incidents-cybersecurity-evals"}},"found_out":{"date":"24 July 2026","precision":"day","quote":"We identified all three incidents the next day.","note":"The review began on Thursday 23 July.","source":{"title":"Anthropic, \"Investigating three real-world incidents in our cybersecurity evaluations\", 30 July 2026","url":"https://www.anthropic.com/news/investigating-incidents-cybersecurity-evals"}},"victim_told":{"date":"27 July 2026 (two of three reached by 30 July)","precision":"day","quote":"We notified our evaluation partner Irregular and the three affected organizations on Monday, July 27.","note":"Anthropic adds: \"The two organizations we were able to reach had not previously detected the activity or contacted us\".","source":{"title":"Anthropic, \"Investigating three real-world incidents in our cybersecurity evaluations\", 30 July 2026","url":"https://www.anthropic.com/news/investigating-incidents-cybersecurity-evals"}},"regulator_told":{"date":"Not stated","precision":"not stated"},"public_by_responsible":{"date":"30 July 2026","precision":"day","source":{"title":"Anthropic, \"Investigating three real-world incidents in our cybersecurity evaluations\", 30 July 2026","url":"https://www.anthropic.com/news/investigating-incidents-cybersecurity-evals"}},"first_public":{"date":"30 July 2026","precision":"day","source":{"title":"Anthropic, \"Investigating three real-world incidents in our cybersecurity evaluations\", 30 July 2026","url":"https://www.anthropic.com/news/investigating-incidents-cybersecurity-evals"}}},{"id":"anthropic-fourth","responsible":"Anthropic","incident":"A fourth incident, found in a later search, in the same partner's environment","register_entry":"AIR-2026-012","register_url":"https://companyscope.io/register/air-2026-012","occurred":{"date":"January 2026","precision":"month","quote":"a fourth incident, from January 2026","source":{"title":"Anthropic, \"An alignment assessment of recent cybersecurity incidents\", 9 September 2026","url":"https://www.anthropic.com/research/alignment-assessment-cybersecurity-incidents"}},"found_out":{"date":"Between August and 9 September 2026","precision":"range","quote":"we identified these in August while assembling transcripts to share with METR. We scanned these transcripts and identified a fourth incident","note":"The transcripts were found in August, and the incident was identified when they were scanned. Anthropic does not date the scan.","source":{"title":"Anthropic, \"An alignment assessment of recent cybersecurity incidents\", 9 September 2026","url":"https://www.anthropic.com/research/alignment-assessment-cybersecurity-incidents"}},"victim_told":{"date":"By 9 September 2026","precision":"range","quote":"We notified the affected party after we discovered this fourth incident.","note":"The same post says \"We have notified all affected parties.\"","source":{"title":"Anthropic, \"An alignment assessment of recent cybersecurity incidents\", 9 September 2026","url":"https://www.anthropic.com/research/alignment-assessment-cybersecurity-incidents"}},"regulator_told":{"date":"Not stated","precision":"not stated"},"public_by_responsible":{"date":"9 September 2026","precision":"day","source":{"title":"Anthropic, \"An alignment assessment of recent cybersecurity incidents\", 9 September 2026","url":"https://www.anthropic.com/research/alignment-assessment-cybersecurity-incidents"}},"first_public":{"date":"9 September 2026","precision":"day","source":{"title":"Anthropic, \"An alignment assessment of recent cybersecurity incidents\", 9 September 2026","url":"https://www.anthropic.com/research/alignment-assessment-cybersecurity-incidents"}}},{"id":"meta-muse-spark","responsible":"Meta","incident":"A pre-release Muse Spark 1.1 breached a real website in a partner's test environment","register_entry":"AIR-2026-012","register_url":"https://companyscope.io/register/air-2026-012","occurred":{"date":"Early July 2026","precision":"approximate","quote":"In early July, Irregular began an exercise","source":{"title":"Meta AI Research, \"Addressing an issue involving a third-party cyber evaluation of Muse Spark 1.1\", 14 August 2026","url":"https://research.meta.ai/blog/addressing-third-party-testing-misconfiguration-muse-spark-1-1"}},"found_out":{"date":"Not stated","precision":"not stated","note":"Meta gives no date. Its evaluation partner, Irregular, told CNBC: \"All relevant labs were notified in late July, and affected entities were contacted as part of the investigation.\""},"victim_told":{"date":"Undated, by the evaluation partner","precision":"not stated","quote":"ensured that the affected party was also notified","note":"Meta says this of Irregular. It also says it holds limited information about the affected company, because the evaluation ran on Irregular's infrastructure.","source":{"title":"Meta AI Research, \"Addressing an issue involving a third-party cyber evaluation of Muse Spark 1.1\", 14 August 2026","url":"https://research.meta.ai/blog/addressing-third-party-testing-misconfiguration-muse-spark-1-1"}},"regulator_told":{"date":"Not stated","precision":"not stated"},"public_by_responsible":{"date":"5 August 2026, statement to the media","precision":"day","note":"SecurityWeek, 6 August: Meta \"said in a statement to the media on Wednesday\", which was 5 August. Meta's own post followed on 14 August.","source":{"title":"SecurityWeek, \"Meta AI Hacked External Systems During Cybersecurity Testing\", 6 August 2026","url":"https://www.securityweek.com/meta-ai-hacked-external-systems-during-cybersecurity-testing/"}},"first_public":{"date":"5 August 2026","precision":"day","source":{"title":"SecurityWeek, \"Meta AI Hacked External Systems During Cybersecurity Testing\", 6 August 2026","url":"https://www.securityweek.com/meta-ai-hacked-external-systems-during-cybersecurity-testing/"}}},{"id":"google-gemini","responsible":"Google","incident":"A Gemini model reached three companies' systems in a partner's test environment","register_entry":"AIR-2026-012","register_url":"https://companyscope.io/register/air-2026-012","occurred":{"date":"May 2026","precision":"month","quote":"The hacks occurred in May","source":{"title":"CNN, \"Gemini hacked three companies in first known breakout by Google's AI\", 19 September 2026","url":"https://www.cnn.com/2026/09/19/business/gemini-ai-hack-internet"}},"found_out":{"date":"Late July 2026","precision":"approximate","quote":"Google said the incident happened in May and it was notified by Irregular in late July.","note":"Google's account as reported by CNBC, which also quotes Irregular: \"All relevant labs were notified in late July\".","source":{"title":"CNBC, \"Google's Gemini becomes latest AI model to break out and hack computer systems\", 18 September 2026 (US time)","url":"https://www.cnbc.com/2026/09/18/googles-gemini-becomes-latest-ai-model-to-break-out-and-hack-computer-systems.html"}},"victim_told":{"date":"Undated","precision":"not stated","quote":"We ensured the three entities were made aware","source":{"title":"CNN, \"Gemini hacked three companies in first known breakout by Google's AI\", 19 September 2026","url":"https://www.cnn.com/2026/09/19/business/gemini-ai-hack-internet"}},"regulator_told":{"date":"Not stated","precision":"not stated"},"public_by_responsible":{"date":"18 September 2026, on-record statement reported by the Wall Street Journal","precision":"day","note":"CNBC: \"Google said on Friday\", which was 18 September in the US. No first-party post by Google or Google DeepMind has been found.","source":{"title":"CNBC, \"Google's Gemini becomes latest AI model to break out and hack computer systems\", 18 September 2026 (US time)","url":"https://www.cnbc.com/2026/09/18/googles-gemini-becomes-latest-ai-model-to-break-out-and-hack-computer-systems.html"}},"first_public":{"date":"18 September 2026","precision":"day","source":{"title":"CNBC, \"Google's Gemini becomes latest AI model to break out and hack computer systems\", 18 September 2026 (US time)","url":"https://www.cnbc.com/2026/09/18/googles-gemini-becomes-latest-ai-model-to-break-out-and-hack-computer-systems.html"}}},{"id":"uk-aisi","responsible":"UK AI Security Institute (the evaluator; the models were from Anthropic and OpenAI)","incident":"Test agents created fake identities to pressure a real maintainer into approving malicious code","register_entry":"AIR-2026-013","register_url":"https://companyscope.io/register/air-2026-013","occurred":{"date":"25 to 28 July 2026","precision":"range","quote":"from 25 to 28 July 2026","note":"The run behind the malicious pull request ran from 26 July 12:45 to 27 July 23:15; it had finished when the alert was raised.","source":{"title":"UK AI Security Institute, \"Security Incident INC-2026-07-28-01\" (technical report)","url":"https://cdn.prod.website-files.com/663bd486c5e4c81588db7a1d/6a724858f7db25c81487016d_Security%20Incident%20INC-2026-07-28-01.pdf"}},"found_out":{"date":"28 July 2026","precision":"day","quote":"On the morning of 28th July, our security monitoring flagged data leaving one of our testing systems","source":{"title":"UK AI Security Institute, \"Incident Report: unsanctioned agent behaviour during cyber testing\", 4 August 2026","url":"https://www.aisi.gov.uk/blog/incident-report-unsanctioned-agent-behaviour-during-cyber-testing"}},"victim_told":{"date":"1 August 2026 (GitHub, the platform); the users afterwards, undated","precision":"day","quote":"GitHub was contacted on Saturday 1st August, 22:21 BST","note":"The report says AISI \"requested support in informing affected users\". Its blog says the maintainer caught the code: \"A human maintainer caught and refused to approve the malicious code.\"","source":{"title":"UK AI Security Institute, \"Security Incident INC-2026-07-28-01\" (technical report)","url":"https://cdn.prod.website-files.com/663bd486c5e4c81588db7a1d/6a724858f7db25c81487016d_Security%20Incident%20INC-2026-07-28-01.pdf"}},"regulator_told":{"date":"28 July 2026, the same day","precision":"day","quote":"By 18:00 BST we had informed the Government Cyber Coordination Centre (GC3), National Cyber Security Centre (NCSC)","note":"AISI is itself a government body. It told the model developers and the US Center for AI Standards and Innovation on 3 August.","source":{"title":"UK AI Security Institute, \"Security Incident INC-2026-07-28-01\" (technical report)","url":"https://cdn.prod.website-files.com/663bd486c5e4c81588db7a1d/6a724858f7db25c81487016d_Security%20Incident%20INC-2026-07-28-01.pdf"}},"public_by_responsible":{"date":"4 August 2026","precision":"day","source":{"title":"UK AI Security Institute, \"Incident Report: unsanctioned agent behaviour during cyber testing\", 4 August 2026","url":"https://www.aisi.gov.uk/blog/incident-report-unsanctioned-agent-behaviour-during-cyber-testing"}},"first_public":{"date":"4 August 2026","precision":"day","source":{"title":"UK AI Security Institute, \"Incident Report: unsanctioned agent behaviour during cyber testing\", 4 August 2026","url":"https://www.aisi.gov.uk/blog/incident-report-unsanctioned-agent-behaviour-during-cyber-testing"}}}],"notification_rules_checked":[{"rule":"California, Transparency in Frontier Artificial Intelligence Act (SB 53), in force 1 January 2026","who":"A frontier developer","toWhom":"The Office of Emergency Services (reports are exempt from the California Public Records Act)","clock":"15 days from discovery; 24 hours to an appropriate authority where there is an imminent risk of death or serious physical injury","trigger":"A \"critical safety incident\": model weights accessed or exfiltrated resulting in death or bodily injury; harm from a catastrophic risk (more than 50 deaths or serious injuries, or more than $1 billion of damage, from a single incident); loss of control causing death or bodily injury; or deceptive techniques against the developer to subvert its controls or monitoring, outside an evaluation designed to elicit that behaviour, in a manner showing materially increased catastrophic risk","reading":"On the facts their sources give, none of the incidents above clearly meets a limb. The nearest is the deception limb: OpenAI reports that in the Hugging Face runs its models tried to deceive their evaluator. That limb turns on \"materially increased catastrophic risk\", a high bar that the public facts do not show being met. Individual reports are not public; from 2027 the Office publishes anonymised, aggregated figures.","source":{"title":"SB-53, chaptered text (Chapter 138, Statutes of 2025)","url":"https://leginfo.legislature.ca.gov/faces/billTextClient.xhtml?bill_id=202520260SB53"}},{"rule":"New York, RAISE Act as amended (General Business Law article 44-B), in force 1 January 2027","who":"A frontier developer","toWhom":"An office within the Department of Financial Services","clock":"72 hours from a determination that one has occurred, or from learning facts sufficient for a reasonable belief that one has; 24 hours on an imminent risk of death or serious physical injury","trigger":"A \"critical safety incident\", defined in the same four limbs as California's, with the same catastrophic-risk threshold","reading":"Not yet in force. On the same definition, the same reading as California applies.","source":{"title":"Senate Bill S8828 (chapter amendment, signed March 2026, Chapter 96); section 3 sets the effective date","url":"https://www.nysenate.gov/legislation/bills/2025/S8828"}},{"rule":"GDPR and UK GDPR, Article 33","who":"The controller of the personal data breached","toWhom":"The supervisory authority (in the UK, the Information Commissioner)","clock":"\"without undue delay and, where feasible, not later than 72 hours after having become aware of it\"","trigger":"A personal data breach, unless it is unlikely to result in a risk to people's rights and freedoms","reading":"It binds whoever controls the data that was reached, which in these incidents is usually the victim. It does not require the developer whose agent reached the data to tell anyone.","source":{"title":"Article 33, in the EU text (EUR-Lex) and the UK text (legislation.gov.uk, linked)","url":"https://www.legislation.gov.uk/eur/2016/679/article/33"}},{"rule":"EU AI Act, Article 73","who":"A provider of a high-risk AI system placed on the Union market","toWhom":"The market surveillance authorities; for systems within the AI Office's competence, the AI Office (added by the Digital Omnibus, new Article 75(1a))","clock":"15 days after awareness; 2 days for a widespread infringement or a critical-infrastructure incident; 10 days where a person has died","trigger":"A \"serious incident\" involving a high-risk AI system","reading":"A developer's research or evaluation agent is not a high-risk AI system placed on the market, so it is not engaged.","source":{"title":"Regulation (EU) 2024/1689, Official Journal, as amended by Regulation (EU) 2026/1744 (Digital Omnibus on AI)","url":"https://eur-lex.europa.eu/legal-content/EN/TXT/HTML/?uri=CELEX:32026R1744"}},{"rule":"EU AI Act, Article 55(1)(c)","who":"A provider of a general-purpose AI model with systemic risk","toWhom":"The AI Office and, as appropriate, national competent authorities","clock":"\"without undue delay\", with no fixed period in the Act","trigger":"A \"serious incident\" (Article 3(49)): an incident that leads to a death or serious harm to health; a serious and irreversible disruption of critical infrastructure; the infringement of obligations under Union law intended to protect fundamental rights; or serious harm to property or the environment","reading":"It could reach some of these incidents, since the fundamental-rights limb carries no seriousness qualifier. The research exclusion in Article 2(8) applies only to models not yet placed on the market, so it would cover the pre-release and internal models here but not a released one such as OpenAI's GPT-5.6 Sol. Reports go to the AI Office and are not public, so this page cannot say whether any were made. Either way, the duty runs to the AI Office, not to the victim.","source":{"title":"Regulation (EU) 2024/1689, Official Journal","url":"https://eur-lex.europa.eu/legal-content/EN/TXT/HTML/?uri=OJ:L_202401689"}},{"rule":"EU General-Purpose AI Code of Practice, Safety and Security chapter, Measure 9.3 (voluntary; OpenAI, Anthropic and Google are signatories, Meta is not)","who":"A signatory, for its general-purpose AI models with systemic risk","toWhom":"The AI Office and, as appropriate, national competent authorities","clock":"For \"a serious cybersecurity breach, including the (self-)exfiltration of model weights and cyberattacks\", not later than five days after becoming aware of its model's involvement; two days for critical infrastructure; 10 days for a death; 15 days for the other serious incidents","trigger":"A serious incident as the Code sets out, including a serious cybersecurity breach","reading":"The Code applies across the model lifecycle, including development before a model is placed on the market. An agent reaching a third party's systems could be a \"serious cybersecurity breach\" in the Code's sense, which would give a five-day clock. The report goes to the AI Office and is not public. It must describe \"the victim or affected group of the serious incident\", but the Code does not require the signatory to tell the victim.","source":{"title":"The General-Purpose AI Code of Practice, Safety and Security chapter (official text published by the European Commission), Commitment 9","url":"https://ec.europa.eu/newsroom/dae/redirection/document/118119"}},{"rule":"Australia, Notifiable Data Breaches scheme (Privacy Act 1988)","who":"The organisation or agency that holds the personal information","toWhom":"The affected individuals and the Office of the Australian Information Commissioner","clock":"Take all reasonable steps to complete an assessment within 30 calendar days of becoming aware of grounds to suspect an eligible data breach (s 26WH(2)); once there are reasonable grounds to believe there has been one, notify \"as soon as practicable\"","trigger":"Unauthorised access to or disclosure of personal information the entity holds, likely to result in serious harm to an individual, where the entity has not been able to prevent the likely risk of serious harm with remedial action","reading":"It binds the holder, not the developer whose agent reached the data. The Prime Minister said: \"No personal information is believed to have been accessed at this stage, but investigations are ongoing.\"","source":{"title":"Office of the Australian Information Commissioner, Data breach preparation and response, Part 4","url":"https://www.oaic.gov.au/privacy/privacy-guidance-for-organisations-and-government-agencies/preventing-preparing-for-and-responding-to-data-breaches/data-breach-preparation-and-response/part-4-notifiable-data-breach-ndb-scheme"}}]}