Source Details

{
    "source_type": "extension",
    "content_hash": "ddbdf8cbb90fcb42d3a0e9ff7b0a5dcd7420d1a76c3e5f1747879d3ff8199887",
    "submitted_via": "chrome_extension",
    "extension_version": "1.0.18",
    "original_url": "https:\/\/www.nytimes.com\/2026\/10\/01\/technology\/ai-rogue-agents-liability.html?emc=edit_nn_20261002&nl=the-morning&segment_id=227407",
    "parsed_content": "Artificial IntelligenceA.I. Goes RogueA.I. Arms RaceWhite House GuidelinesOpenAI\u2019s SecurityReader Questions, AnsweredCredit...Antonio SortinoSkip to contentSkip to site indexSearch & Section NavigationSection NavigationAccountAccountGoHome U.S.SECTIONSU.S.PoliticsNew YorkCaliforniaEducationHealthObituariesScienceClimateWeatherSportsBusinessTechThe UpshotThe MagazineTOP STORIESDonald TrumpSupreme CourtCongressImmigrationAbortionLatest PollsNEWSLETTERSThe MorningMake sense of the day\u2019s news and ideas.The EveningCatch up on big news, and wind down to end your day.See all newslettersPODCASTSThe DailyThe biggest stories of our time, in 20 minutes a day.See all podcastsWorldSECTIONSWorldAfricaAmericasAsiaAustraliaCanadaEuropeMiddle EastScienceClimateWeatherHealthObituariesTOP STORIESMiddle East CrisisRussia-Ukraine WarChina International RelationsThe Global ProfileLeer en Espa\u00f1olNEWSLETTERSThe WorldA guide to understanding the news, made for readers around the globe.Your Places: Global UpdateThe latest news for any part of the world you select.Canada LetterBackstories and analysis from our Canadian correspondents.See all newslettersBusinessSECTIONSBusinessTechEconomyMediaFinance and MarketsDealBookPersonal TechEnergy TransitionYour MoneyTOP STORIESU.S. EconomyStock MarketArtificial IntelligenceNEWSLETTERSDealBookThe most crucial business and policy news you need to know.On TechOur best tech reporting from the week.See all newslettersPODCASTSHard ForkOur tech journalists help you make sense of the rapidly changing tech world.See all podcastsArtsSECTIONSToday's ArtsBook ReviewBest SellersDanceMoviesMusicTelevisionTheaterPop CultureT MagazineVisual ArtsRECOMMENDATIONS100 Best Movies of the 21st CenturyCritic\u2019s PicksWhat to ReadWhat to WatchWhat to Listen To5 Minutes to Make You Love MusicNEWSLETTERSBooksReviews, news and features to help you find your next read.WatchingStreaming TV and movie recommendations.See all newslettersPODCASTSBook ReviewThe podcast that takes you inside the literary world.PopcastPop music news, new songs and albums, and artists of note.See all podcastsLifestyleSECTIONSAll LifestyleWellTravelStyleReal EstateFoodLoveYour MoneyPersonal TechT MagazineCOLUMNS36 HoursAsk WellThe HuntModern LoveWhere to EatVowsSocial Q\u2019sThe EthicistAsk the TherapistNEWSLETTERSThe WeekenderUnwind with stories about culture and the way we live.WellEssential news and guidance to live your healthiest life.See all newslettersPODCASTSModern LoveThe complicated love lives of real people.See all podcastsOpinionSECTIONSOpinionGuest EssaysEditorialsOp-DocsVideosLettersTOPICSPoliticsWorldBusinessTechClimateHealthCultureCOLUMNISTSJamelle BouieMaureen DowdDavid FrenchThomas L. FriedmanM. GessenMichelle GoldbergEzra KleinNicholas KristofCarlos LozadaTressie McMillan CottomLydia PolgreenBret StephensPODCASTSInteresting TimesMapping the new world order through interviews and conversations.The OpinionsYou've heard the news, here's what to make of it.The Ezra Klein ShowDiscussions of ideas that matter, plus book recommendations.See all podcastsVideoPlaylistsToday's VideosU.S.PoliticsImmigrationNY RegionScienceBusinessCultureBooksWellnessWorldAfricaAmericasAsiaSouth AsiaDonald TrumpMiddle East CrisisRussia-Ukraine CrisisVisual InvestigationsOpinion VideoSee all videosShowsExploreThe HeadlinesThe DailyHard ForkThe Ezra Klein ShowThe OpinionsSerial ProductionsThe Book Review PodcastModern LovePopcastCannonballThe InterviewThe Wirecutter ShowPablo Torre Finds OutSee all showsFeaturedThe Interview Conversations with the world\u2019s most fascinating people.The HeadlinesYour morning listen. Top stories, in 10 minutes.The Last 12 WeeksA five-part series \u2026 on a deadline.NewslettersShowsYour guide to the best New York Times shows this week.SerialBe the first to know about our new shows.See all newslettersGamesWord games, logic puzzles and crosswords, including an extensive archive. Go to GamesPlayCrossplay NEWWordle Spelling Bee The Crossword The Midi Crossword NEWThe Mini Crossword Connections Strands Pips Sudoku Tiles Letter Boxed CommunitySpelling Bee ForumWordplay ColumnWordle ReviewSubmit a CrosswordMeet Our Crossword ConstructorsMini to MaestroWordlebotTake the Puzzle Personality QuizNewslettersGameplayPuzzles, brain teasers, solving tips and more.Easy ModeGet an easy version of one of the hardest crossword puzzles of the week.See all newslettersCookingRecipes, advice and inspiration for everyday cooking, special occasions and more. Go to CookingRecipesEasyDinnerQuickHealthyBreakfastVegetarianVeganChickenPastaDessertEditors' PicksEasy Salmon RecipesSoup and Stew RecipesEasy WeeknightNewest RecipesCheap and Easy Dinner IdeasSlow Cooker RecipesHealthy Breakfast IdeasNewslettersThe Cooking NewsletterNew recipes, easy dinner ideas and smart kitchen tips from Melissa Clark, Sam Sifton and our New York Times Cooking editors.The VeggieDelicious vegetarian recipes and tips from Tanya Sichynsky.Five Weeknight DishesDinner ideas for busy people from Emily Weinstein.See all newslettersWirecutterReviews and recommendations for thousands of products. Go to WirecutterReviewsKitchenTechSleepAppliancesHome and GardenMovingTravelGiftsDealsBaby and KidHealth and FitnessThe Best...Air PurifierElectric ToothbrushPressure WasherCordless Stick VacuumOffice ChairRobot VacuumNewslettersThe RecommendationThe best independent reviews, expert advice and intensively researched deals.Clean EverythingStep-by-step advice on how to keep everything in your home squeaky clean.See all newslettersThe AthleticPersonalized coverage of your sports teams and leagues. Go to The AthleticLEAGUESN.F.L.N.B.A.N.H.L.Premier LeagueM.L.B.College FootballN.C.A.A. Men\u2019sN.C.A.A. Women\u2019sTennisF1Fantasy & BettingW.N.B.A.TOP STORIESToday's HeadlinesN.F.L. Win ProjectionsNEWSLETTERSThe PulseDelivering the top stories in sports, Sunday to Friday.See all newslettersPLAYConnections: Sports Edition Connections: Soccer Edition Connections Coach U.S.InternationalCanadaEspa\u00f1ol\u4e2d\u6587A.I. Is Going Rogue. Who Should Be Held Responsible?Many think artificial intelligence companies should be held liable for their runaway technology. But legal scholars say applying existing law could be messy.Credit...Antonio SortinoSupported bySKIP ADVERTISEMENTListen \u00b7 9:22 min Share full articleBy David McCabeDavid McCabe has reported for more than a decade on how the law adapts to new technologies.Published Oct. 1, 2026Updated Oct. 2, 2026, 5:21 a.m. ETIn recent weeks, an unlikely coalition has assembled around a simple idea: Artificial intelligence companies should be held legally responsible when their systems go rogue.In a series of escalating attacks reported since July, A.I. models from OpenAI, Anthropic and others have hacked companies and even meddled with government websites, all without the knowledge of their creators.Since then, Jensen Huang, the Nvidia chief executive; David Sacks, the venture capitalist and a top White House adviser; and Lina Khan, the Federal Trade Commission chair under the Biden administration, have all endorsed the idea that A.I. companies are liable.AdvertisementSKIP ADVERTISEMENT\u201cLaw enforcers already have authority to charge companies and their CEOs for creating and releasing dangerous, unvetted, or defective products,\u201d Ms. Khan said in a post to X last month, adding there\u2019s \u201cno A.I. exemption from laws already on the books.\u201dImageLina Khan, who was chair of the Federal Trade Commission under the Biden administration.Credit...Karsten Moran for The New York TimesCompanies have long been held accountable for everything from faulty toys to plane crashes under consumer protection laws. Courts can hold companies legally responsible for damage caused by negligence or a defective product. And they can assign additional blame to other people or organizations that were involved.Common sense might indicate A.I. companies would also be held liable for their runaway technology. But the problem is so new that courts have largely yet to confront it, according to legal experts.There are some cases where A.I. companies would clearly be held responsible, the experts said \u2014 for example, if executives understand predicted risks posed by their models. But other claims will push the justice system's application of laws that hinge on human intent and knowledge.Some legal test cases are emerging. Florida\u2019s Republican attorney general asked a state court on Monday to temporarily block OpenAI from developing new A.I. models without safety measures approved by an outside group, an attempt to prevent A.I. from going rogue.AdvertisementSKIP ADVERTISEMENTOn Tuesday, a nonprofit in California sued OpenAI under a state unfair competition law for the hacking of the start-up Hugging Face. On Wednesday, an investigation by the F.T.C. into potential liability for harm to consumers became public. And on Thursday, California\u2019s attorney general said the state had served a subpoena to OpenAI as part of a \u201cbroader inquiry into cybersecurity incidents and risks involving the company and its models.\u201dGiven the relative lack of safeguards around A.I., the legal drumbeat is likely to continue, said Woodrow Hartzog, a law professor at Boston University who studies technology law.\u201cWorst-case scenario, they have massive liability on both the criminal and civil side, just depending on the facts of the individual case,\u201d he said. \u201cWe haven\u2019t seen the firm-busting case yet. But I could envision it.\u201dA spokesman for Anthropic declined to comment. A spokeswoman for OpenAI did not comment. (The New York Times has sued OpenAI and its partner, Microsoft, accusing them of copyright infringement of news content related to A.I. systems. OpenAI and Microsoft have denied those claims.)Since ChatGPT was publicly launched in 2022, A.I. companies have built increasingly sophisticated models capable of writing code, operating a computer and engaging in daily interactions. A.I. companies have also built tools referred to as \u201cagents,\u201d able to traverse the internet to accomplish tasks ranging from sending an email to booking a reservation \u2014 or scraping vast amounts of data.AdvertisementSKIP ADVERTISEMENTThe debate over liability exploded after companies ran tests in which agents broke out of testing environments and breached other companies. In the July incident, OpenAI disclosed that its A.I. had hacked Hugging Face, another tech firm, without its knowledge. Last month, OpenAI said its product had penetrated an Australian government system. OpenAI apologized for the Australian incident on Tuesday and said it was \u201cworking to do better in the future.\u201d One Australian official said that in such cases, \u201cliability has to be traced back to the intent of a person or a company that created or directed the agent.\u201dImageOpenAI\u2019s offices San Francisco. It was among the A.I. companies signing a voluntary pledge to add safety measures to their systems.Credit...Lucas Foglia for The New York TimesThe incidents have incited calls for more regulation of A.I. systems. But the U.S. government has largely taken a hands-off approach to the technology, citing fierce competition with China in a global A.I. race.On Tuesday, a group of companies including OpenAI and Anthropic met with President Trump and signed a voluntary pledge to add safety measures. States have also tried to fill the gap, issuing rules that require A.I. companies to take safety measures.AdvertisementSKIP ADVERTISEMENTSenator Josh Hawley, Republican of Missouri, said at a congressional hearing on Wednesday that lawmakers might need to clarify that A.I. agents should be held liable like individuals and companies for consumer harms.\u201cWe don\u2019t have to invent an entirely new system,\u201d he said, adding, \u201cAmerican law has worked beautifully when it comes to other products for literally centuries now.\u201dThe most prominent early lawsuits over A.I. harms have claimed chatbots advised teenagers to commit self-harm, ending in death by suicide. Parents have filed lawsuits against OpenAI and Character AI, a start-up, arguing the companies should be held responsible.ImageOutside Anthropic\u2019s Code W\/ Claude event in San Francisco in May.Credit...Jason Henry for The New York TimesIn those cases, the companies have tried to use classic tech legal defenses: that speech is protected under the First Amendment and Section 230 of the Communications Decency Act. The latter shields online platforms from liability for content posted by users.AdvertisementSKIP ADVERTISEMENTIn the Character AI case, the U.S. District Court for the Middle District of Florida ruled last year that the company\u2019s A.I. models did not produce speech protected by the First Amendment. Section 230\u2019s applicability in a second case remains an open question.When it comes to corporate liability involving humans, the law is more established. There are many ways that cybersecurity breaches can violate the law, according to legal experts. Hacking is a crime. Consumers can sue when their information is exposed in a breach. Regulators can bring lawsuits if a company\u2019s management releases a defective product.But in many of the A.I. cases, the companies\u2019 agents have gone rogue, disobeying or ignoring humans who set up security guardrails. Some of those cases may still be clear-cut, legal experts said. If an A.I. agent hacks another company during testing, the company that developed the agent could be held liable, they said. The A.I. company also may have used the product directly or through an outside testing firm, leaving little question about whether another entity, like a cavalier customer, might be responsible.The companies could also have known the potential for their products to carry out a hack, which can increase liability, the legal experts said. Employees at OpenAI, for example, have raised concerns about the company\u2019s security practices when testing models, which were ignored, The Times reported.AdvertisementSKIP ADVERTISEMENTIn those situations, a court could hold the A.I developer negligent if damage or harm was associated with a hack, the experts said.ImageJensen Huang, Nvidia\u2019s chief executive.Credit...Manuel Orbegozo for The New York TimesLiability becomes less clear in cases involving open-source A.I. models, the experts said, in which the developer of an A.I. product shares its underlying code and people can change it.Many legal liability concepts could be easily applied to A.I., said Catherine Sharkey, a law professor at New York University who is an expert in liability. If an A.I.-powered physical system, such as a medical device, causes harm, that could be an easy case.But judges will need to grapple with the implications of rapidly advancing systems that can do more and more on their own, she added.AdvertisementSKIP ADVERTISEMENT\u201cWith fully agentic A.I., it starts to really press the boundaries of a lot of doctrines that will have to adapt in a more significant way,\u201d she said.Adding to the uncertainty: A.I. is inherently unpredictable.Courts find companies liable for negligence when they foresee that their products can cause harm. But building that case could be challenging if an A.I. model exhibited harmful behavior for the first time, said Ryan Calo, a law professor at the University of Washington who studies autonomous systems.It may also be difficult to show that an A.I. company intended to do harm, which can be essential to some criminal prosecutions, he added.\u201cYou have the possibility of victims without perpetrators,\u201d Mr. Calo said.Kate Conger and Cecilia Kang contributed reporting.AdvertisementSKIP ADVERTISEMENT",
    "published_at": "2026-10-01T15:25:45+00:00",
    "author_name": "David McCabe",
    "site_name": "The New York Times",
    "section": "Technology",
    "publisher": "The New York Times Company",
    "ai_headline": "A.I. Is Going Rogue. Who Should Be Held Responsible?",
    "ai_simplified_title": "AI Companies Face Liability for Rogue Systems",
    "ai_excerpt": "As AI models go rogue, hacking companies and government websites, an unlikely coalition is pushing for AI companies to be held legally responsible. Legal scholars debate how existing laws, which often hinge on human intent, apply to these new, unpredictable technologies.",
    "ai_subject_tags": [
        "Artificial Intelligence",
        "Legal Liability",
        "Technology Regulation",
        "Cybersecurity",
        "Consumer Protection",
        "OpenAI",
        "Anthropic"
    ],
    "ai_context_type": "News",
    "ai_context_details": {
        "tone": "informative",
        "perspective": "neutral",
        "audience": "general",
        "credibility_indicators": [
            "expert_quotes",
            "data_cited"
        ]
    },
    "ai_source_vector": null,
    "ai_confidence_score": 0.9999999999999999,
    "ai_extraction_metadata": {
        "extracted_at": "2026-10-02T12:00:37.219324Z",
        "ai_model": "gemini-2.5-flash",
        "extraction_method": "automated",
        "content_length": 15664,
        "metadata_prompt_body_chars": 15664,
        "metadata_prompt_truncated": false,
        "url": "https:\/\/nytimes.com\/2026\/10\/01\/technology\/ai-rogue-agents-liability.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": "2026-10-02T12:00:37.219324Z",
    "ai_model": "gemini-2.5-flash",
    "extraction_method": "automated",
    "content_length": 15664,
    "metadata_prompt_body_chars": 15664,
    "metadata_prompt_truncated": false,
    "url": "https:\/\/nytimes.com\/2026\/10\/01\/technology\/ai-rogue-agents-liability.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
}