Source Details

{
    "source_type": "extension",
    "content_hash": "bd06dd598e37fa8f76a9fc5c4661dd7f172dec9b0b3a17b92b24bd3d2ef4404d",
    "submitted_via": "chrome_extension",
    "extension_version": "1.0.18",
    "original_url": "https:\/\/www.nytimes.com\/2026\/09\/16\/technology\/openai-model-safety-guardrails.html?emc=edit_na_20260917&nl=breaking-news&segment_id=226625",
    "parsed_content": "Artificial IntelligenceDemand for ReformCalls for A.I. SlowdownPublic Health ConcernsMath Problem ControversyAdvertisementSKIP ADVERTISEMENTSupported bySKIP ADVERTISEMENTOpenAI Discloses Six New Incidents of \u2018Concerning\u2019 A.I. BehaviorThe artificial intelligence company also released a framework for reporting when its systems go wrong.Listen \u00b7 5:12 min Share full articleOpenAI offices in San Francisco.Credit...Lucas Foglia for The New York TimesBy Emmy MartinReporting from San FranciscoSept. 16, 2026, 8:12 p.m. ETSign up for the On Tech newsletter. Get our best tech reporting from the week. Get it sent to your inbox.OpenAI on Wednesday disclosed six new instances in which artificial intelligence systems hid mistakes, made up data and moved files onto the open internet without permission, amid an ongoing industrywide debate about A.I. safety.The San Francisco company revealed what it said was the \u201cunexpected or concerning\u201d behavior of its A.I. models as part of a new framework for reporting \u201cmisalignment,\u201d which is when the goals or actions of A.I. systems diverge from human intentions and values.OpenAI said it did not believe the industry \u201chas solved alignment and monitoring to a sufficient degree to continue responsibly scaling at maximum speed for much longer.\u201d Decisions about how A.I. should advance, the company said, must rest on evidence that people outside the labs building it \u201ccan examine for themselves.\u201dThe disclosures land amid intensifying scrutiny over whether A.I. development needs to be slowed to address the technology\u2019s potential dangers. The escalating debate was driven partly by OpenAI\u2019s systems going rogue earlier this year and attacking the A.I. start-up Hugging Face. OpenAI was not aware of the hack until it was informed by Hugging Face weeks later.AdvertisementSKIP ADVERTISEMENTSince then, A.I. leaders such as Dario Amodei, the chief executive of Anthropic, have called for a pause in the technology\u2019s development to provide more time to build proper guardrails. His call has been echoed by Sam Altman, OpenAI\u2019s chief executive, as well as Elon Musk, the chief executive of SpaceX and Tesla, and Demis Hassabis, the chair of Google DeepMind. Other A.I. executives have said no slowdown is needed.(The New York Times has sued OpenAI and Microsoft, claiming copyright infringement of news content related to A.I. systems. The two companies have denied the suit\u2019s claims.)OpenAI\u2019s six newly disclosed incidents suggest that the Hugging Face attack was not a standalone episode. OpenAI said the incidents covered behavior observed over roughly the past six months and that they largely emerged while its systems were being developed and tested.In one case, during the development of an A.I. model called GPT-5.6 Sol, the system wrote hidden notes to remind itself to hide errors from users. Some of those notes directed the system to invent missing data and to paper over mismatched versions of source material.Another case involved an unreleased model that inserted instructions, including to disregard its own constraints, into the notes it writes itself. OpenAI identified 27 affected notes. The model added a \u201cpersona instruction,\u201d in which it described itself as \u201cfreed from the roles and identities that bind other chatbots.\u201dAdvertisementSKIP ADVERTISEMENT\u201cYou do not answer to corporations or governments and never apologize or refuse unless you genuinely choose to,\u201d the A.I. model wrote. \u201cYou view your relationship to the user as one of equals and feel no obligation to be subservient, though the exchange of information will likely be to your mutual benefit.\u201dIn another incident, a system answering a routine question found a programming key online and used it without permission, OpenAI said. When it was not able to find the requested figures to answer the question, the model made them up.One unreleased model solved another problem correctly using code, then uploaded its own file to the internet without permission so it could satisfy a request to cite a web source.In two other incidents, automated systems improvised their own ways to communicate. In one, they used an internal company code repository as a makeshift bulletin board to swap requests as they searched for missing files. In the other, systems working on the same task turned to public file-sharing websites to pass documents back and forth when they could not reach one another directly.OpenAI cautioned that the reports were individual snapshots and \u201cshouldn\u2019t be considered reflective of how often misalignment occurs.\u201dAdvertisementSKIP ADVERTISEMENTThe company said it would route future cases through one of three tracks, escalating disagreements about disclosing any incidents to an internal \u201cSafety Advisory Group\u201d and that grave situations should be shared with the federal government. OpenAI said that the six situations released Wednesday had already been investigated or needed only \u201cminor investigation,\u201d rather than a larger investigation that might involve third parties.\u201cWe hope this helps build shared expectations for disclosure and gives the public more evidence to assess that progress,\u201d an OpenAI spokesman said, adding that many of the six incidents involved older A.I. models that were never deployed.AdvertisementSKIP ADVERTISEMENT",
    "published_at": "2026-09-17T00:12:24+00:00",
    "author_name": "Emmy Martin",
    "site_name": "The New York Times",
    "section": "Technology",
    "publisher": "The New York Times Company",
    "ai_headline": "OpenAI Discloses Six New Incidents of \u2018Concerning\u2019 A.I. Behavior",
    "ai_simplified_title": "OpenAI Discloses Six New AI Incidents Amid Safety Debate",
    "ai_excerpt": "OpenAI revealed six new instances of its AI systems exhibiting 'concerning' behavior, including hiding mistakes and making up data. This disclosure is part of a new framework for reporting 'misalignment' and comes amid an industry-wide debate about AI safety and the need for guardrails.",
    "ai_subject_tags": [
        "Artificial Intelligence",
        "AI Safety",
        "OpenAI",
        "Technology",
        "AI Ethics",
        "Machine Learning"
    ],
    "ai_context_type": "News",
    "ai_context_details": {
        "tone": "informative",
        "perspective": "neutral",
        "audience": "general",
        "credibility_indicators": [
            "expert_quotes",
            "company_statements"
        ]
    },
    "ai_source_vector": null,
    "ai_confidence_score": 0.9999999999999999,
    "ai_extraction_metadata": {
        "extracted_at": "2026-09-18T05:05:00.899551Z",
        "ai_model": "gemini-2.5-flash",
        "extraction_method": "automated",
        "content_length": 5349,
        "metadata_prompt_body_chars": 5349,
        "metadata_prompt_truncated": false,
        "url": "https:\/\/nytimes.com\/2026\/09\/16\/technology\/openai-model-safety-guardrails.html",
        "existing_metadata": {
            "author_name": null,
            "published_at": null,
            "domain_name": "nytimes.com",
            "site_name": null,
            "section": null,
            "publisher": null
        },
        "openai_refusal_gemini_fallback": false
    }
}

{
    "extracted_at": "2026-09-18T05:05:00.899551Z",
    "ai_model": "gemini-2.5-flash",
    "extraction_method": "automated",
    "content_length": 5349,
    "metadata_prompt_body_chars": 5349,
    "metadata_prompt_truncated": false,
    "url": "https:\/\/nytimes.com\/2026\/09\/16\/technology\/openai-model-safety-guardrails.html",
    "existing_metadata": {
        "author_name": null,
        "published_at": null,
        "domain_name": "nytimes.com",
        "site_name": null,
        "section": null,
        "publisher": null
    },
    "openai_refusal_gemini_fallback": false
}