**
In a dramatic turn of events, leading artificial intelligence firm Anthropic has suspended public access to its latest AI models, Fable 5 and Mythos 5, in response to a US government export control directive linked to national security concerns. The company described the situation as a “misunderstanding” and is working diligently to restore access, which it believes has been unjustly revoked.
Government Directive Triggers Model Suspension
On Friday, Anthropic announced the abrupt withdrawal of its cutting-edge AI offerings following the issuance of a directive from the US government. This order, which cited potential security risks, was delivered without specific details, leaving the company to grapple with the implications. Anthropic indicated that the government raised concerns about a possible method to circumvent safeguards intended to prevent the models from being misused, particularly in identifying software vulnerabilities.
This move comes at a time when relations between Anthropic and the US government appeared to be improving, particularly after a prolonged dispute regarding the military’s request to utilise Anthropic’s AI technologies for domestic surveillance and autonomous weapon systems. Following the company’s refusal to comply, it found itself on a government supply chain blacklist, set to be enforced later this year.
Tensions Rise Between AI Developers and Regulators
The suspension of Fable 5 and Mythos 5 highlights the escalating tensions between AI developers and regulatory bodies as the latter seek to impose stricter controls on artificial intelligence technologies. Historically, US export controls have concentrated on the hardware and tools that underpin AI systems, rather than directly limiting access to the AI models themselves.
Anthropic has publicly disagreed with the government’s stance, asserting that the evidence provided regarding the so-called “jailbreak” was vague and insufficient to warrant such drastic action. The firm stated, “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.” This clash underscores the complexities surrounding the regulation of rapidly evolving AI technologies and the different perceptions of risk involved.
Implications for the AI Landscape
Interestingly, just days prior to the directive, Anthropic had advocated for increased government oversight of AI systems, including the authority to prohibit models deemed too risky. However, the company contended that the recent directive was not in line with principles of fair regulation. The Pentagon’s chief information officer, Kirsten Davies, expressed support for prioritising national security over commercial interests, stating, “Some things are simply more important than revenue cycles, clickbait, and pre-IPO valuation. America First. Always.”
The timing of this directive coincides with the recent launch of Claude Fable 5, part of a new generation of AI models designed with strict guardrails to restrict their application in high-risk areas, such as cybersecurity. Despite these precautions, experts have warned that misuse of such models could lead to an increase in sophisticated cyberattacks, particularly in sectors reliant on outdated technology.
A Call for Clarity and Fairness
Anthropic maintains that it has collaborated with the US government to establish safety measures prior to the launch of Fable 5, asserting that similar capabilities exist in models from rival AI companies. The company has emphasised the need for clarity, stating, “If this standard was applied across the industry, we believe it would essentially halt all new model deployments for all frontier model providers.”
Amazon Web Services (AWS), Anthropic’s cloud service provider, confirmed that the company had requested the revocation of access to its models for all users globally. A government official corroborated that the directive includes a suspension of access for foreign nationals, raising questions about the future of key personnel within Anthropic, many of whom are non-US citizens.
Why it Matters
This episode illustrates the growing divide between technological innovation and regulatory oversight in the AI sector. As governments grapple with the potential risks associated with AI, companies like Anthropic are caught in a precarious position, navigating the complexities of compliance while striving to push the boundaries of technological advancement. The outcome of this situation could set significant precedents for the future of AI development, particularly in how access and utilisation are governed on a global scale. The implications extend beyond Anthropic, potentially affecting the entire landscape of artificial intelligence and its integration into society.