US blocks India, others from accessing Anthropic’s top AI models
text_fields
Washington: The United States has imposed restrictions on access to Anthropic’s latest artificial intelligence models, Claude Fable 5 and Claude Mythos 5, for foreign users, including those in India, citing national security concerns. The export-control order was issued by the US government on June 12.
The restrictions affect Anthropic’s most advanced AI systems. Following the order, the company suspended access to Claude Fable 5 and Claude Mythos 5 for users outside the United States.
According to reports, US intelligence and security agencies raised concerns that the models could be used to identify software vulnerabilities and assist in cyberattacks targeting banks, government networks and other critical infrastructure. The action reportedly followed research indicating that certain safety mechanisms within the models could be bypassed to uncover security flaws.
The US Department of Commerce subsequently moved to restrict access to the systems. The measure marks a significant expansion of US technology controls, which have previously focused mainly on semiconductors and AI hardware.
Anthropic has opposed the restrictions, stating that the models underwent extensive testing by independent researchers and AI safety organisations before their release. The company maintains that similar capabilities are available in other publicly accessible AI models and has begun discussions with US authorities seeking restoration of access.
Reports have also linked the decision to disagreements between Anthropic and the Trump administration. The company had previously indicated that it would not provide its AI systems for certain domestic surveillance applications or fully autonomous weapons programmes.
The move has prompted discussion in countries that depend heavily on foreign AI technologies. Zoho founder Sridhar Vembu said the development highlights the need for India to strengthen domestic AI capabilities and invest in open-source alternatives. Industry observers view the restrictions as a signal that the United States may increasingly extend technology controls beyond hardware to advanced software and AI systems.