Washington: Amazon CEO Andy Jassy reportedly joined discussions with senior officials in the Trump administration regarding potential security concerns linked to Anthropic’s latest artificial intelligence models, according to a report citing sources familiar with the matter.
The development comes shortly after AI startup Anthropic unexpectedly suspended access to its newest AI systems worldwide following directives issued by the U.S. government on national security grounds.
U.S. Government Flags Security Concerns
Anthropic revealed that federal authorities had identified a possible method to bypass certain safety protections built into its advanced AI models. According to the company, officials expressed concerns that users could potentially exploit the models to discover cybersecurity vulnerabilities.
As a result, the government directed the company to impose restrictions on access to its latest AI offerings, known as Fable 5 and Mythos 5.
Anthropic Halts Global Access
In response to the order, Anthropic announced that it would disable access to the affected models globally. The San Francisco-based company stated that the identified weakness involved a “jailbreak” technique that could circumvent some safety mechanisms.
However, Anthropic maintained that the issue exposed only limited vulnerabilities and argued that similar capabilities exist in other publicly available AI systems.
Amazon’s Role Comes Into Focus
Reports indicate that Andy Jassy was among technology executives who discussed the matter with government officials during the week. While Amazon did not directly confirm its involvement in conversations related to Anthropic’s AI products, the company acknowledged that governments frequently seek its input on technology and security matters.
An Amazon spokesperson stated that it is common for public authorities to consult major cloud providers regarding potential security risks but declined to comment on specific discussions.
Export Controls at the Center of the Dispute
Anthropic said the restrictions were implemented under U.S. export-control regulations. The move reportedly prevents foreign nationals from accessing the company’s most advanced AI models, regardless of whether they are located inside or outside the United States.
The decision has sparked debate within the technology and policy communities, particularly because the restrictions extend beyond rival nations and affect citizens of allied countries as well.
Experts Question Broad Restrictions
Several experts who generally support tighter controls on advanced AI technologies expressed surprise at the scope of the measures.
Critics argue that the rules could limit research and development activities involving foreign employees working for U.S.-based AI companies. Some analysts believe the restrictions may have unintended consequences for international collaboration in artificial intelligence research.
The latest developments highlight the growing tension between rapid AI innovation and national security concerns, as governments worldwide seek to determine how advanced AI systems should be regulated and controlled.
