OKX Restricts Hong Kong Employees from Using Anthropic AI Model

Crypto exchange OKX has restricted employees in Hong Kong from using Anthropic's AI model Claude, following a brief corporate account suspension. This move comes amidst growing US-China tensions over artificial intelligence.

The company's corporate account was briefly suspended, and as a result, OKX stopped providing access to Claude for employees in Hong Kong and those traveling through China in early August. Anthropic's policies prohibit access from unsupported locations, and currently, its supported locations do not include Hong Kong or mainland China.

OKX has been pushing employees to use AI, with AI proficiency becoming part of employee performance evaluations. The company spends between $6 million and $8 million per month on large-language-model providers. Employees were previously able to use both Claude and OpenAI's ChatGPT.

This is not an isolated incident; Goldman Sachs faced a similar situation earlier this year when its employees in Hong Kong lost access to Claude. Chinese technology giant Alibaba took the opposite approach, banning its own employees from using Claude Code due to high-risk software concerns.