Two of Wall Street's biggest banks have cut their Hong Kong employees off from Anthropic's Claude AI, a sign of how US-China tensions are reshaping which technology global firms can use where.
According to the Financial Times, JPMorgan Chase has stopped its Hong Kong staff from accessing Anthropic's AI, removing Claude from an internal list of approved large language models available to employees. The FT reports the move was driven by Anthropic's licensing terms.
JPMorgan is not acting alone. Reporting carried by MSN and others says JPMorgan is following Goldman Sachs, which had already restricted Claude for staff in the territory. Both banks cited usage terms and potential national security concerns, and the restrictions follow a US government order — part of what Storyboard18 describes as a broader US crackdown.
The distinction that matters here is geographic. The reporting frames this as a restriction on employees outside the US — specifically in Hong Kong — rather than a blanket ban across each bank. That points to the underlying issue: rules governing how American AI tools can be used in jurisdictions tied to China.
The timing adds another layer. As one report framed it, Anthropic faces more scrutiny ahead of its IPO. Restrictions imposed by marquee financial clients like JPMorgan and Goldman are the kind of detail investors and regulators tend to watch closely as a company prepares to go public.
Why it matters: when the world's largest banks start drawing borders around which AI their staff can use, it signals that cutting-edge AI tools are now caught up in the same geopolitical fault lines as chips and other strategic technology.