JPMorgan blocked Claude access for Hong Kong staff, according to reports citing the Financial Times, adding another flashpoint to the fight over advanced AI tools in global finance.
The move matters because banks are racing to use AI internally while navigating licensing terms, national-security pressure and different rules across China, Hong Kong, Europe and the United States.
What happened
The Financial Times reported that JPMorgan Chase stopped Hong Kong employees from accessing Anthropic's AI models. The Straits Times, Investing.com and Moneycontrol summarized the report, noting that Goldman Sachs had made a similar move earlier.
The reported restriction was tied to Anthropic usage terms and the availability of Claude models through internal approved-tool lists.
Why JPMorgan Claude access matters
For banks, AI access is no longer just an IT decision. It affects analyst productivity, compliance workflows, cross-border operations and the ability of global teams to use the same tool set.
Market impact
The direct stock impact may be limited, but the strategic impact is larger. If frontier AI access fragments by region, global banks may need parallel systems, higher compliance spending and slower AI rollouts.
Key numbers
- Straits Times published its report on June 18, 2026.
- The report said JPMorgan removed Claude from an internal approved-model list for Hong Kong staff.
- Goldman Sachs reportedly made a similar Hong Kong access move earlier.
- Anthropic's Claude is one of the leading enterprise AI model families.
- The issue sits at the intersection of bank compliance, AI licensing and geopolitics.