JPMorgan blocks Anthropic AI access for Hong Kong staff, FT reports

JPMorgan’s reported move to block Anthropic’s AI models for Hong Kong staff highlights intensifying scrutiny of how U.S. AI technology is used outside the United States. Reuters reported on June 18 that, following wording in Anthropic’s licensing terms with the bank, JPMorgan removed Claude models from an internal menu of approved large-language-model options available to employees in Hong Kong. The Financial Times cited three people familiar with the matter. Goldman Sachs took a similar step in April, removing Claude from tools approved for its Hong Kong-based bankers. The actions come as U.S.-China tensions rise over AI access, data security, and advanced computing. Reuters noted that neither JPMorgan nor Anthropic commented, and it could not independently verify the report.