How Claude Is Set to Power Core Banking, Airline, and Insurance Systems

Hello, I'm Ai Learning's Lao Zhang. I recently saw news that Anthropic and DXC Technology have announced a multi‑year global strategic alliance to embed Claude into the core systems of the world’s largest banks, airlines, insurers, and government agencies—not just as a chatbot or marketing tool, but deep within the infrastructure that handles transfers and ticketing.

This Is Worth Discussing

DXC Technology, though not a household name, employs 115,000 people across 70 countries and maintains the core IT systems for the biggest financial institutions, airlines, and insurers. Because its clients operate in heavily regulated sectors, choosing an AI partner is extremely cautious; Claude’s selection indicates it has passed the strictest enterprise security and compliance tests.

95% of the Code Is Written by Claude – A Crazy Figure

In April, DXC launched the AI‑native operations platform DXC OASIS, and according to DXC, more than 95% of its code is authored by Claude. This is a production‑grade platform already serving over 50 enterprise customers, not a demo or internal tool. DXC claims Claude accelerates software development by tenfold. The combination of 95% AI‑generated code and a ten‑times speed boost shows Claude is acting as the primary developer, while human engineers focus on code review and architectural decisions.

For technologists, this suggests future software teams may consist of a few architects, many AI agents, and review engineers, turning coding from a core human skill into an AI‑driven operation.

"Forward‑Deployed Engineer" Model – Learning from Palantir

DXC plans to train tens of thousands of "Claude‑certified Forward‑Deployed Engineers" (FDEs) who will embed directly within client organizations. This mirrors Palantir’s classic approach, which used FDEs to win contracts with the U.S. government and military; DXC and Anthropic aim to apply the same strategy to global finance and aviation.

These engineers will first earn Anthropic Academy certification, then complete DXC‑specific courses tailored to core systems, learning to use Claude and then apply it within banking core systems.

Four Entry Points of the Alliance

The alliance’s first phase focuses on four domains:

Insurance : Using Claude’s agent capabilities to modernize decades‑old COBOL‑based insurance core systems—a monumental engineering effort.

Modernization‑as‑a‑Service (MaaS) : Analyzing, refactoring, and modernizing legacy codebases for enterprise customers, leveraging Claude’s leading code‑understanding abilities.

Cybersecurity : Building a "never‑down" AI security sub‑agent with Claude Security to monitor anomalies 24/7 in DXC’s Security Operations Center.

Application Services : Embedding Claude directly into DXC’s operational applications to handle routine maintenance and management, effectively handing over repetitive ops work to AI.

Claude’s Enterprise Footprint Is Expanding Rapidly

Beyond DXC, Claude’s partner ecosystem already includes major players such as Notion, Figma, Asana, Stripe, Square, Plaid, Cloudflare, Vercel, Slack, HubSpot, Intercom, Atlassian, Linear, Sentry, and Microsoft. With the Model Context Protocol (MCP) connector suite, Claude is positioned as the "central nervous system" of the enterprise software ecosystem, aiming to become a standard AI infrastructure rather than just a model provider.

Anthropic Academy – Building a Certification System

Anthropic has launched Anthropic Academy, offering a full curriculum and certifications ranging from API development to enterprise deployment, including Claude Code hands‑on, Cowork basics, and MCP courses. Establishing a certification program signals Anthropic’s transition from a product company to a platform company, similar to AWS or Salesforce.

My Takeaways

The alliance sends three clear signals:

First, AI is entering the "deep water" of core business systems. Shallow chatbot and content‑generation use cases are past; the next frontier is decades‑old, mission‑critical infrastructure.

Second, Anthropic’s enterprise strategy is crystal clear. While OpenAI focuses on consumer‑facing ChatGPT and Google pursues an integrated AI suite, Anthropic is targeting heavily regulated, high‑value enterprise sectors, potentially building the deepest moat.

Third, the "95% AI‑generated code" milestone should alert every developer. It doesn’t mean programmers will disappear, but the most valuable skills will shift toward code review, architecture, and decision‑making.

Conclusion

The DXC + Anthropic alliance demonstrates that Claude is ready for the world’s most stringent and critical technology environments. This is not a pilot or a proof‑of‑concept; it is a multi‑year contract backed by tens of thousands of certified engineers and a proven platform (OASIS). For anyone tracking AI’s evolution, this development carries more weight than any benchmark score.