\n But companies building AI models like Claude and ChatGPT face a more immediate and prosaic risk: They\u2019re likely to get sued. A lot.\n <\/p>\n
\n AI companies have so far avoided major lawsuits over recent incidents in which they\u2019ve failed to keep AI agents \u2014 systems capable of working autonomously with little oversight \u2014 contained to their testing environments. But some legal experts and policymakers say it\u2019s pretty clear: The companies can and should be held responsible\n <\/p>\n
\n \u201cThe question of \u2018how is the law going to apply to this?\u2019 is a lot easier to answer if you see (AI) as it is, and not as this hypothetical thing that many in the industry are proposing it might someday be,\u201d said Kristen Thomasen, associate professor at University of Windsor and an expert in robotics law. \u201cThe law has dealt with new technologies for as long as it has existed.\u201d\n <\/p>\n
\n The AI industry often refers to the breakout incidents<\/a> as the technology going \u201crogue\u201d \u2014 language that some critics say is designed to amplify the companies\u2019 claims that their technology is so powerful that even its creators struggle to contain it.\n <\/p>\n \n It may also be a tactic to try to shield themselves from responsibility and future legal claims, Thomasen said.\n <\/p>\n \n Still, she said, the liability questions around AI models have a simple answer: \u201cIt\u2019s a thing that this company is putting out into the world. If it does harm, there are laws that regulate that.\u201d\n <\/p>\n \n So far, the real-world harms done by rogue bots have been far from calamitous but nonetheless costly for the organizations they\u2019ve infiltrated. Last week, the nonprofit that operates Wikipedia said<\/a> it had found evidence of unauthorized OpenAI agents attempting to make \u201cpotentially malicious edits\u201d to some pages and pummeling it with so much traffic that they may have triggered a partial outage.\n <\/p>\n \n The investigation \u201cstopped work inside key parts of the organization for us to focus on this,\u201d Eric Mill, product manager for safety and security at Wikimedia, told CNN. \u201cWe put our very best people on it\u2026 and it\u2019s a pretty messy process.\u201d\n <\/p>\n \n Wikimedia launched its investigation this summer as global alarm grew over the Hugging Face attack<\/a> and dozens of other<\/a> instances of internal AI models breaking out of their testing environments to hack other computer systems. Meta<\/a> and Anthropic<\/a> also disclosed that their test models have hacked into other systems without authorization.\n <\/p>\n \n The Wikimedia probe is significant in part because of what didn\u2019t happen: There was no catastrophic, existential threat to humanity of the sort that AI executives often allude to. But the cost \u2014 more than two weeks of work carried out by the leading engineers at a global, $200 million-plus organization \u2014 was real.\n <\/p>\n \n Wikimedia isn\u2019t pushing any one policy or technological solution, and Mill said the organization is in communication with OpenAI about its findings. (OpenAI didn\u2019t respond to CNN\u2019s request for comment.) But the nonprofit hopes these incidents are a wake-up call for AI developers.\n <\/p>\n \n \u201cAI companies are not doing enough to secure their systems and protect the public from the harm they cause,\u201d Wikimedia said in its blog post<\/a> last week. \u201cThat burden is falling onto everyone else, including smaller organizations.\u201d\n <\/p>\n \n Anthropic, which is preparing to go public next month in one of the largest IPOs in history, disclosed in a regulatory filing that because the legal framework around AI is \u201cunsettled,\u201d<\/a> those agents\u2019 actions could expose the company to \u201csignificant and unpredictable\u201d lawsuits, according to a copy of the document viewed by Reuters.\n <\/p>\n \n The potential litigation onslaught isn\u2019t just a headache for company lawyers or a quandary for academics \u2014 it could also become a major drag on Anthropic\u2019s bottom line. Litigation costs are a financial drain for any company, but it could become especially problematic for Anthropic, which is burning through cash faster than it can bring it in. The company tore through $42 billion last year while bringing in just $4.6 billion in revenue, according to the document seen by Reuters.\n <\/p>\n \n The CEOs of both Anthropic and OpenAI have argued for new regulations tailored to the industry. Dario Amodei recently penned an essay<\/a> arguing that companies \u201cpace the frontier\u201d and foster international cooperation on safety protocols.\n <\/p>\n \n But the rival AI firms have diverged slightly on the liability question.\n <\/p>\n \n OpenAI CEO Sam Altman earlier this year favored an Illinois bill <\/a>that would limit when AI labs can be held liable, even when their products cause \u201ccritical harm.\u201d (Anthropic came out against it.)\n <\/p>\n \n And while Altman endorsed Amodei\u2019s \u201cpace the frontier\u201d proposal, he also told POLITICO<\/a> earlier week that his worldview isn\u2019t perfectly aligned with his rival\u2019s: OpenAI believes \u201cthe world should accept some bad things happening for the benefits of this technology,\u201d he said. Altman also told the news site<\/a> that \u201cif something goes wrong with our models during training, there\u2019s gonna be some version of that we need to be responsible for.\u201d\n <\/p>\n \n An Anthropic spokesperson told CNN the company sees liability laws as a backstop, rather than a substitute, for safety regulations<\/a>. Anthropic has never proposed exemptions from existing liability laws, the spokesperson added.\n <\/p>\n \n Policymakers on the left and the right have sought to cut through some of the more nuanced questions of future regulation and underscore the legal liabilities AI is already subject to.\n <\/p>\n \n Lina Khan, the former Federal Trade Commission chair under President Joe Biden, has become a prominent challenger to AI executives, arguing on X that \u201cwe shouldn\u2019t let discussions about new legal regimes distract from the fact that there\u2019s no AI exemption from laws already on the books.\u201d\n <\/p>\n \n Meanwhile, Republican Senator Josh Hawley, alongside Democratic counterpart Chris Murphy, introduced a bill last week that seeks to ensure \u201cAI agent operators and developers are held liable for hacking incidents.\u201d\n <\/p>\n![\"Anthropic](\"https:\/\/www.newsbeep.com\/nz\/wp-content\/uploads\/2026\/10\/20260803-dario-amodei-sam-altman-split.jpg\")
<\/p>\n