US Government Forces Anthropic to Shut Down Fable 5 and Mythos 5 Models

Title: AI Accountability Tightens as Governments, Courts Act

01US Government Forces Anthropic to Shut Down Fable 5 and Mythos 5 Models

The US government has issued an unprecedented order requiring Anthropic to disable access to its most powerful AI models, citing national security concerns. According to reporting by Wired, the government became aware of a potential method for "jailbreaking" Fable 5, which could bypass safety guardrails. The order applies to all foreign nationals worldwide, including Anthropic's own international employees, effectively shutting down the models for all customers to ensure compliance.

Amazon CEO Andy Jassy reportedly raised concerns about the models with US officials before the crackdown took effect. As TechCrunch reported, Jassy's communications with government authorities may have triggered the action. The move marks a significant escalation in government oversight of frontier AI capabilities.

Anthropic has publicly disputed the decision. "We disagree that the finding of a narrow potential jailbreak should be cause for recalling a commercial model deployed to hundreds of millions of people," the company stated in a blog post cited by TechCrunch. The company called the directive "deeply disappointing" given its history of proactive safety work and transparency.

02Court Rules Google Liable for False AI Overviews

A landmark court ruling has established that companies operating AI systems assume legal liability for damages caused by false statements their AI generates. According to Wired, the ruling holds that a company "that designs, trains, operates, and manages an AI system must assume legal liability for any damages caused by the responses it generates." This sets a major precedent for AI accountability at a time when AI-generated content is increasingly integrated into consumer products.

The case centered on Google's AI Overviews feature, which generates summarized responses to search queries. Courts found that Google bore responsibility for inaccurate or misleading information produced by the system. The ruling could have far-reaching implications for the entire AI industry, potentially exposing companies to lawsuits whenever their AI systems generate harmful or deceptive content.

The decision represents a shift from previous legal frameworks that often shielded technology companies from liability for third-party content. Now, AI companies that actively shape their systems' outputs through training and design may face direct legal consequences for those outputs. Industry observers expect this ruling to trigger a wave of similar litigation against AI providers.

03State Attorneys General Launch Investigation Into OpenAI

Multiple state attorneys general have launched an investigation into OpenAI's business practices, signaling escalating regulatory scrutiny of major AI companies. According to TechCrunch, the investigation covers a broad range of issues, including OpenAI's advertising policies and its handling of health data. The states involved have not been publicly identified, but the scope of inquiry suggests a comprehensive review of the company's operations.

The investigation adds to mounting pressure on OpenAI from regulators worldwide. As AI systems become more integrated into daily life, authorities are increasingly focused on how these companies collect, use, and protect sensitive information. Health data handling is particularly sensitive, given AI systems' ability to infer personal details from seemingly innocuous inputs.

The development comes amid broader crackdowns on the AI industry. Combined with the Anthropic shutdown and the Google liability ruling, the investigation suggests regulators are moving from theoretical concerns to concrete enforcement actions. AI companies may need to significantly expand their compliance and legal teams as they face a new era of accountability.

Also today

04KPMG pulls report on AI usage due to apparent hallucinations — The consulting firm withdrew an AI-related report after discovering the AI system had generated unreliable or fabricated information.
05Police officer investigated for using AI to 'create evidence' in multiple cases — A Derbyshire police officer is under investigation for allegedly using AI to fabricate evidence in multiple criminal cases.
06Meta's months-old AI unit is a soul-crushing gulag, say the engineers stuck inside it — A report claims Meta's 6,500-person AI division is experiencing internal revolt amid reports of poor working conditions.
07Chinese cybercrime operation that used AI to scam 'hundreds of thousands of victims' sued by Google — Google filed suit against a Chinese group called Outsider Enterprise for using AI to send 2.5 million scam text messages to hundreds of thousands of victims.
08Israeli firm BlackCore suspected of meddling in New York and Scotland votes — An Israeli firm faces suspicion of using AI to interfere with elections in New York, Scotland, and France.
09PeopleSoft 0-day affecting hundreds of organizations steals gigabytes of data — A critical zero-day vulnerability in Oracle's PeopleSoft software is enabling attackers to steal large amounts of data from hundreds of organizations.