Anthropic faces scrutiny over data privacy and intelligence sharing policies

Reports indicate that several U.S. companies, including Nvidia, have begun restricting or halting the use of Anthropic's frontier models. The shift follows Anthropic's unilateral updates to its privacy policy, which now mandates a 30-day retention period for user interaction data for security reviews without an opt-out option. An analysis of 13 revisions to the company's privacy policy since 2023 suggests increasing risks to user data security. Unlike many AI firms that only provide data to law enforcement under legal requirements, Anthropic's policy allows the company to share user data with U.S. intelligence agencies at its own discretion, without legal proceedings.