trending
Barclays is extending its use of Anthropic’s Claude AI models across software development, customer support and operational processes as the bank looks to modernise legacy technology and improve efficiency.
Barclays Expands Claude AI Use
The lender expects Claude Code to reach around 50% of its developer population by the end of 2026, with adoption expanding to the majority of software engineers during 2027.
The move forms part of a wider technology programme in which Barclays has said AI could contribute to approximately £2 billion in efficiency savings.
AI Moves Deeper Into Core Operations
Barclays is already using Claude beyond software engineering.
Its Colleague Knowledge Assistant, introduced in 2025, helps employees retrieve information more quickly when supporting customers. More than 16,000 staff have adopted the tool, which has processed over one million searches.
The bank is also using Claude within its Global Markets business to classify, enrich and route incoming emails.
That system handles around 120,000 messages a day, reducing manual processing while helping staff identify and respond to relevant client requests more efficiently.
Legacy Modernisation Becomes a Key Use Case
The expansion of Claude Code is particularly significant for Barclays’ technology estate.
Large banks continue to operate substantial volumes of legacy infrastructure, making software modernisation costly and resource-intensive.
Barclays says Claude is being used to automate routine development tasks, improve software quality and support engineers working on older systems.
That could allow technical teams to focus on more complex transformation work rather than repetitive coding and maintenance.
Anne Marie Darling, Barclays’ group co-chief operating officer, said the objective is not simply AI adoption, but changing how work is performed across the organisation.
The broader significance is that generative AI is moving beyond isolated banking pilots.
For large financial institutions, the next phase is increasingly about embedding AI into software development, operations and customer servicing at scale, while maintaining the governance and human oversight required in a regulated environment.
Comments