DXC to embed Claude in banks’ and airlines’ core systems

TL;DR:

• Anthropic and IT services firm DXC Technology have announced a multi-year alliance to bring Claude into the mission-critical systems DXC runs for banks, airlines, insurers and governments.
• DXC will train tens of thousands of Claude-certified “forward-deployed engineers” embedded inside customer organisations.
• The firm says Claude wrote more than 95% of the code for its new OASIS managed-services platform, now serving over 50 customers.

Anthropic is pushing deeper into the regulated enterprise, announcing a multi-year global alliance with DXC Technology — one of the largest IT services companies — to embed its Claude models in the systems that run banks, airlines, insurers and government agencies. The deal turns DXC’s decades of operating those systems into a distribution channel for frontier AI.

From its own operations outward

The structure follows a now-familiar “proved it on ourselves first” playbook. DXC says it used Claude across its own operations — 115,000 employees in 70 countries — before offering it to clients, and that Claude generated more than 95% of the code for DXC OASIS, its AI-native orchestration platform, with engineers reviewing the output. OASIS launched in April and now serves over 50 customers, with Claude as its default model.

The alliance starts in four areas: insurance modernisation, refactoring legacy codebases (“Modernization as a Service”), an always-on cybersecurity subagent built on Claude Security, and application maintenance. DXC will certify engineers through Anthropic Academy, layering its own curriculum on top, and joins the Claude Partner Network.

As a vendor announcement, the eye-catching figures — tens of thousands of engineers, a tenfold development speed-up — are claims rather than independently verified outcomes, and “more than a dozen” pilots have a way of outrunning production reality. Still, the direction matters for UK readers: it puts AI agents inside the regulated banking and insurance systems where errors carry real consequences, echoing the caution behind UK regulators’ warnings on frontier AI and bank cyber resilience and Anthropic’s earlier Project Glasswing cyber-defence work with BT.

Looking forward

Embedding Claude in core financial and aviation systems is a bigger trust ask than coding assistants or chatbots. Whether DXC’s compliance-first framing holds up under audit — and under the first high-profile failure — will shape how fast other regulated sectors follow.