Indian Ministries Halt Deployment Of Foreign AI Models For Cybersecurity

Indian government ministries have implemented a temporary suspension on the use of advanced foreign AI models for critical cybersecurity operations. This directive specifically impacts high-end models such as OpenAI's GPT-5.6 and Anthropic's Claude Fable 5 or Mythos 5. The decision serves as a precautionary measure to ensure that sensitive national cyber infrastructure remains under strict domestic oversight while the government formulates comprehensive regulatory frameworks.

The move is driven by significant concerns regarding data management, accessibility, and national security. A pivotal factor in this decision was an incident in June 2026, where Anthropic briefly suspended access to its models due to US export controls. This event highlighted the risks of relying on foreign-controlled AI infrastructure for essential national tasks, prompting officials to prioritize the development of sovereign AI capabilities that are immune to external geopolitical shifts.

The government intends to use this pause to establish definitive protocols for AI usage in sensitive sectors. By creating a controlled environment, authorities aim to mitigate risks associated with third-party model dependencies. This initiative underscores a broader national strategy to foster indigenous AI development, ensuring that India's critical digital assets are protected by technology that aligns with national security interests and long-term strategic autonomy in the rapidly evolving global AI landscape.