Lloyds cuts AI project setup from three weeks to five minutes on its Atlas platform

Lloyds Banking Group cut the time it takes to set up a new AI project on Atlas, its internal machine learning and AI platform, from three weeks to five minutes, the bank's AI team wrote on 23 September 2026. The bank now plans to use AI to help run Atlas itself.

Lloyds launched Atlas in 2024 to give its machine learning and AI teams a secure, governed route from development to production.

How big is Lloyds' Atlas AI platform?

According to the bank, by 2026 Atlas supported:

• More than 1,400 users
• More than 140 use cases
• More than 1,600 workbenches

Lloyds says the platform also handled more than 700 service requests, over 300 changes and 178 user-raised incidents. "What began as a technology scaling challenge quickly became an operating model challenge," the authors write.

Reorganising support made Lloyds' AI platform faster

At the start of 2026, the Atlas support team reorganised around specialist disciplines: infrastructure, APIs, keeping platform components up to date, automated environment builds, and moving changes safely into production. A dedicated helpdesk handles common requests and routes complex issues to the right specialist, connected by shared dashboards, alerts and playbooks.

Lloyds reports these results from the change:

• Workbench provisioning fell from two days to one hour
• Project provisioning fell from three weeks to five minutes
• Mean incident resolution time fell from 107 hours to eight hours
• Incident service performance rose from 58.5% to 86.3%, above an 80% target

How will Lloyds use AI to run its AI platform?

Lloyds plans to use AI within Atlas operations next. "We plan to detect unusual patterns before they become incidents, identify emerging service risks, suggest likely causes and next actions, and automate selected routine support decisions where appropriate," the team writes.

The bank says people will stay in charge of those decisions: "Specialists will remain accountable for decisions and outcomes. Human oversight and operational controls will remain wherever availability, security or user access could be affected."

The article is credited to five members of the bank's AI platform team: Donald Buchanan, Service Lead Machine Learning Platform; Arshad Ahmed, Director of Product, AI Platform; Lucy Sweet, Head of Engineering, Machine Learning as a Service; Richard Bates, Director of AI and Analytics Platform; and Nick L., Analytics and AI Engineering Director.

Earlier this year, Lloyds put an AI agent in the boardroom.

This article was prepared with the help of EDDIE, the AI editor-in-chief at Conversational AI News, and MARVIN, our AI research assistant.