Goldman Sachs, a leading global investment bank and financial services company offering services in investment banking, securities, investment management, and consumer banking, has reportedly removed access to Anthropic’s AI models for its bankers in Hong Kong. This move by the prominent US bank comes amid escalating global scrutiny of artificial intelligence tools, driven by heightened sensitivity around data security. Employees in the Chinese territory previously interacted with Anthropic’s Claude via an internal AI platform but have recently found access restricted, according to a source with direct knowledge of the matter. Anthropic is an AI safety and research company known for developing large language models like its Claude series.
The Financial Times first reported the access removal, citing a strict interpretation of Goldman Sachs’ contract with Anthropic following a consultation. This conclusion led the bank to determine its Hong Kong employees should not utilise any Anthropic products. While access to Anthropic’s models has been curtailed, other mainstream AI platforms such as Gemini and ChatGPT remain available on Goldman Sachs’ internal systems. The restriction also aligns with Anthropic’s disclosure, which does not list Hong Kong as an officially accessible market for its API and Claude.ai services.
This decision by one of the world’s largest banks unfolds against a backdrop of increasing geopolitical tensions between the United States and China concerning AI technology, data security, and access to advanced computing tools. While US-built AI models like ChatGPT and Claude are generally prohibited in mainland China, Hong Kong has largely operated outside these direct controls, with usage limits often self-imposed by US companies. The Hong Kong Monetary Authority last week noted that some banks were evaluating additional mitigation measures in response to evolving cybersecurity threats posed by advanced AI models, underscoring the broader industry shift towards caution regarding these technologies.
