Anthropic is expanding access to its powerful new Claude AI model to British financial institutions within days, despite warnings from senior finance leaders about its risks. The tool was previously limited to US firms like Amazon, Apple, and Microsoft.
UK banks will gain access to Anthropic's latest Claude model next week, marking the first expansion of a tool the company initially deemed too risky for public release.
The move raises concerns among global finance leaders, who have flagged potential stability risks from deploying advanced AI systems in critical financial infrastructure. Some executives worry the model's capabilities could amplify existing vulnerabilities in markets already facing pressures from geopolitical tensions and energy shocks.
Anthropologic has maintained strict controls over the model's rollout. The AI tool is currently available only to select US enterprises including Amazon, Apple, and Microsoft. The decision to extend access to British institutions suggests confidence in managing deployment, though transparency around specific safeguards remains limited.
Financial sector regulators have not publicly commented on the expansion. The timing coincides with broader industry concerns about AI integration in banking, where systems influence trading, risk assessment, and customer decisions at scale.
Anthropics has not disclosed specific limitations or guardrails being implemented for financial sector users. Industry observers note the contrast between the company's cautious initial rollout and the accelerating pace of institutional access.
The expansion reflects competitive pressure in the AI sector, where companies balance safety protocols with market demands for advanced capabilities. Other AI developers have faced similar tensions between controlled releases and commercial growth.
Stakeholders in UK finance have not yet publicly responded to the imminent access. Conversations among banking executives suggest mixed reactions—some view the tool as essential for competitive advantage, while others remain skeptical about integration timelines and risk management.
Z.ai released GLM-5.3's weights on Hugging Face under a new license that requires large companies to undergo security review before hosting the model. The change marks a departure from the standard MIT license.
Anthropic has introduced the Model Hardware Standard (MHS), a unified interface enabling AI agents to operate robotic arms, lab instruments, and other physical devices. Early testing shows integration time has dropped from weeks to hours.
Open-weight AI companies—those releasing freely available models—are attracting major acquisition interest from tech giants. The trend reflects growing capital investment in the business model of distributing AI models at no cost.
Google Deepmind has upgraded its Co-Scientist AI system to autonomously plan experiments, operate lab equipment, and publish scientific papers. The Gemini-based multi-agent platform demonstrated experimentally validated results across materials science, chemistry, and medical AI development.