Summary
The U.S. government has ordered Anthropic to immediately suspend global access to its latest Claude Fable 5 and Mythos 5 models, citing national security concerns. The move has triggered intense debate about AI sovereignty in countries like India and Canada, while raising fresh security concerns for the DeFi industry, which has already lost over $840 million this year.
Emergency Directive Shakes Global AI Industry
At 5:21 PM Eastern Time on June 13, AI safety company Anthropic received an emergency directive from the U.S. government requiring it to immediately suspend global access to its two newest AI models, Claude Fable 5 and Mythos 5. While the export control order nominally targets foreign nationals, in practice Anthropic was forced to shut off access for all users worldwide, including its own foreign national employees.
Anthropic confirmed compliance with the government directive on social media platform X, but made clear it believes the government's decision was wrong. In a blog post, the company argued that the capabilities apparently causing government concern are already available in other publicly accessible models. This incident represents not just another friction point between technology companies and government regulators, but has sparked a global conversation about AI sovereignty, cybersecurity, and technological dependence.
Mythos 5: Too Powerful to Release
Mythos 5 is Anthropic's most capable AI model to date, first previewed by the company in early April. However, due to its exceptional ability to find security vulnerabilities in software, Anthropic has kept it tightly restricted. According to company disclosures, Mythos identified flaws in every major operating system and web browser it tested.
Given the potential risks of such capabilities, Anthropic did not release Mythos 5 broadly. Instead, it launched a controlled program called Project Glasswing, sharing the model with roughly 50 vetted organizations, including tech giants Amazon, Apple, Google, Microsoft, and CrowdStrike, specifically for defensive cybersecurity work.
Fable 5 represented Anthropic's answer to obvious commercial pressure, a version of Mythos fitted with guardrails that block responses in high-risk areas like cybersecurity and biology. The company argued this made it safe enough for general release. Released just three days before the government order, Fable 5 immediately became the most capable AI model available to the public, according to benchmark tests from Vals AI, a company that tracks AI technology performance.
Amazon's Role and Questions of Conflict
The Wall Street Journal reported that Amazon CEO Andy Jassy may have been the source of security concerns that led to the government action. According to the report, Jassy told Treasury Secretary Scott Bessent and other government officials that Amazon researchers used Claude Fable 5 to obtain information that could be used in cyberattacks. The government subsequently imposed the export control ban on Fable 5 and Mythos 5.
This role has raised questions about potential conflicts of interest, as Amazon is a major Anthropic investor. An Amazon spokesperson said in a statement that while it is not uncommon for governments to seek counsel on potential security risks, the company does not share the details of those discussions. The spokesperson also pointed to an update stating that AWS has been affected by the model cutoff.
David Sacks, Trump's former AI czar who now co-chairs the President's Council of Advisors on Science and Technology, offered his own account, claiming that a highly credible trusted partner of both Anthropic and the U.S. government came forward with a jailbreak. Sacks added that the administration asked Anthropic CEO Dario Amodei to fix the jailbreak or de-deploy the model, but Dario refused.
Canadian Prime Minister's 2008 Financial Crisis Analogy
Canadian Prime Minister Mark Carney told reporters during a visit to Ireland that the U.S. export ban forcing Anthropic to shut down Fable 5 and Mythos 5 demonstrates the danger of depending on a small number of powerful AI models. Carney framed the suspension as a warning about systemic vulnerability rather than a failure by any single company.
Carney stated that the situation we are in collectively right now with Mythos and Fable is something that can happen with over-reliance on certain models. Nobody has done anything wrong in this situation, but we will have done something wrong if we just accept this, do not take the lesson, and do not build out and diversify.
The former central banker then drew a direct parallel to the 2008 financial crisis. Carney, who led both the Bank of Canada and the Bank of England before entering politics, said we have similar things in terms of model risk and called for redundancy and diversity in AI infrastructure, the same principles regulators imposed on the banking system after the collapse of Lehman Brothers.
The analogy carries particular weight because of who is making it. Carney was governor of the Bank of Canada during the 2008 crisis and subsequently became the first non-British governor of the Bank of England, where he spent six years strengthening financial system resilience. When he warns that concentrated AI dependence mirrors the systemic linkages that nearly destroyed the global banking system, the comparison draws on direct experience rather than rhetorical convenience.
India's AI Sovereignty Awakening
Anthropic's sudden move has raised fresh questions across the global technology industry. In India, the decision has reignited a long-running debate over whether one of the world's largest AI markets can afford to rely on technologies built and controlled elsewhere.
The timing was particularly delicate. The announcement came shortly after Anthropic announced a partnership with Indian IT services giant Tata Consultancy Services to expand enterprise AI adoption in India. The government directive requiring suspension of access to Fable 5 and Mythos 5 for all foreign nationals, including Anthropic's own foreign national employees, underlined how closely India's AI ambitions have become tied to technologies developed and governed in the U.S.
The development has triggered debate among Indian founders, investors, and policy experts over whether the country should accelerate efforts to build domestic AI capabilities, deepen investment in open-source alternatives, or continue relying on a handful of U.S. frontier model providers. For some, the episode is a wake-up call on technological dependence. For others, it is a reminder that access to increasingly critical AI systems can be suspended.
Many in India's tech community view this incident as the strongest argument yet for sovereign AI development. The experience of suddenly losing access to the most advanced models in a market that had just announced a major partnership with a global leading AI company reinforces the urgency of building indigenous capabilities.
New Threats Facing the DeFi Industry
Anthropic's new Claude Fable 5 model offers users access to stronger, faster reasoning and coding capabilities, landing in a crypto market beset by security problems and potentially exacerbating them.
The company released Claude Fable 5 on Tuesday as the first public model in the Mythos class and, Anthropic says, its most powerful yet. So powerful, in fact, the company released two versions: one for widespread use and another restricted to vetted security users.
Security experts warn that advanced AI will not invent fundamentally new crypto hacks, but will dramatically speed up finding misconfigurations and constructing ways of exploiting weaknesses such as social engineering, exposed keys, and flawed signing flows.
This year's largest DeFi losses, totaling more than $840 million, have mostly stemmed from human error and operational failures rather than smart-contract bugs. Against this backdrop, the emergence of AI tools capable of rapidly identifying system vulnerabilities poses an unprecedented threat to the entire decentralized finance ecosystem.
The concern is not that AI will discover entirely new categories of vulnerabilities, but rather that it will compress the timeline for exploiting known weakness patterns. What might take a human security researcher days or weeks to identify and weaponize could potentially be accomplished by an advanced AI system in hours or minutes. For DeFi protocols operating with billions of dollars in total value locked, this acceleration of the attack cycle represents a fundamental shift in the threat landscape.
Balancing Technological Sovereignty and Security
This incident reveals multiple tensions in the global AI ecosystem: friction between technology companies and government regulators, competition and cooperation among major tech firms, technological dependence between developed and emerging markets, and the tradeoff between AI capability advancement and security risks.
Anthropic's argument that the capabilities of government concern already exist in other publicly available models suggests selective and potentially inconsistent regulatory action. Meanwhile, The Information and Reuters reported that the White House is unlikely to extend similar restrictions to other AI companies and is privately blaming Anthropic's handling of alleged jailbreak vulnerabilities.
For industries relying on these models, from fintech to DeFi, this incident serves as a warning to pay attention not only to technological capability advancement but also to the geopolitical and regulatory risks of accessing those capabilities. In an increasingly interconnected yet increasingly fragmented world, technological sovereignty is no longer an abstract policy goal but a practical business continuity issue.
The episode also highlights the complex interplay between private sector innovation, national security concerns, and international competitiveness. When a government can effectively shut down global access to a cutting-edge AI model with a single directive, it raises fundamental questions about the architecture of AI development and deployment. Should critical AI infrastructure be concentrated in the hands of a few companies in a single jurisdiction? What are the implications for countries and industries that have built dependencies on these systems?
For the digital asset industry specifically, which has long championed decentralization and resistance to single points of failure, the centralization of advanced AI capabilities presents a philosophical as well as practical challenge. The industry's response to this challenge, whether through investment in open-source AI alternatives, development of indigenous capabilities, or diversification of model dependencies, may shape its resilience and autonomy in the years ahead.
Looking Ahead: Lessons for a Fragmented AI Future
The Anthropic model suspension may prove to be a watershed moment in the evolution of global AI governance. It demonstrates that even the most advanced AI capabilities can be subject to sudden access restrictions based on national security determinations. It shows that major investors in AI companies may play complex and potentially conflicting roles in security discussions with governments. And it reveals that countries and industries heavily dependent on a small number of frontier AI models face systemic risks analogous to those that brought down the global financial system in 2008.
As AI capabilities continue to advance and become more deeply embedded in critical infrastructure, from financial systems to cybersecurity defenses, the questions raised by this incident will only become more pressing. The path forward likely requires a combination of technological diversification, regulatory clarity, international coordination, and continued investment in both safety research and indigenous AI capabilities across different jurisdictions.
For now, the immediate impact is clear: Anthropic's most powerful models are offline globally, India's AI partnership ambitions have suffered a setback, Canada's prime minister is calling for AI infrastructure diversification, and the DeFi industry faces the prospect of AI-accelerated attacks with or without access to the models designed to defend against them. How governments, companies, and industries respond to these challenges will shape the AI landscape for years to come.
Source: link