Thought Machine and AWS target Tier 1 bank mainframe debt with AI

A joint Vault Forge and AWS Transform pipeline converts COBOL mainframe logic into cloud-native products, compressing multi-year migrations into days.

By Disrupts Editorial Team Thu, 15 October 2026 AI level 2 of 5

AI level 2 of 5: drafted by our AI editorial assistant from the company's announcement; fact-checked, edited and signed off by Mark Walker, Editorial Director. What the levels mean

A grey, multi-wheeled military air defense missile system with multiple missile launch tubes is displayed in a brightly lit studio with a reflective floor and light gray walls.

Thought Machine, the cloud-native core banking vendor, and Amazon Web Services have launched a joint AI-powered migration pipeline aimed squarely at the legacy mainframe problem that has paralysed Tier 1 banks for decades. The product, Vault Forge powered by Amazon Bedrock, combines AWS Transform's mainframe reverse-engineering tooling with Thought Machine's AI agent suite to extract business logic from COBOL systems and re-synthesise it as simulation-tested Python financial products, all within the bank's own secure cloud environment.

The partnership is significant less for its technical novelty than for the institutional weight behind it. Thought Machine counts JPMorgan Chase, Lloyds, Standard Chartered, and Deutsche Bank among its clients; Deutsche Bank recently signed a 10-year agreement. AWS brings the hyperscaler muscle and the regulatory credibility that cautious risk committees require. Together, they are positioning the solution as the first pipeline to shift the industry's migration approach from laborious line-by-line code translation towards automated business-rule extraction at scale.

From years to days: what the pipeline actually does

The four-stage process begins with AWS Transform performing discovery and reverse engineering on legacy applications, producing structured specifications in Easy Approach to Requirements Syntax (EARS). Vault Forge then orchestrates specialised AI agents via Amazon Bedrock to consolidate redundant legacy product variants, synthesise SDK-compliant Python code and test suites, and deliver fully validated migration proof early in the programme. A human-in-the-loop review gate gives compliance and audit teams explicit sign-off at each stage before any code is promoted to a live sandbox.

Paul Taylor, CEO of Thought Machine, said the partnership "fundamentally rewrites the economics of core modernisation," enabling banks to "deploy AI to rapidly convert complex legacy mainframe specifications into simulation-tested, cloud-native financial products."

Thought Machine is accompanying the launch with a Core Modernisation Accelerator programme, embedding technical architects and forward-deployed engineers inside bank teams to manage governance and architecture from day one, a recognition that technology alone rarely clears the organisational and regulatory hurdles that derail large migration programmes.

The convergence angle: where AI infrastructure meets financial sovereignty

The broader significance of this launch sits at the intersection of AI infrastructure spend, financial regulatory risk, and sovereign competitiveness. Central banks and prudential regulators across the UK, EU, and Asia-Pacific have spent the past three years tightening operational resilience rules. Basel-aligned frameworks increasingly treat unresolved legacy technical debt as a systemic risk factor, not merely an IT cost line. That regulatory pressure is now colliding with the availability of frontier AI tooling capable of actually tackling the problem at scale, creating a narrow but commercially enormous window.

For hyperscalers, core banking is one of the last enterprise verticals where cloud penetration remains structurally low because migration risk has always outweighed migration benefit. A credible AI-assisted path changes that calculus, and AWS's commitment here reflects the strategic importance of winning Tier 1 financial institution workloads as cloud infrastructure competition with Microsoft Azure and Google Cloud intensifies.

The capital implications extend further. Banks that can compress multi-year, nine-figure modernisation programmes carry lower operational risk premiums and free up technology budgets previously locked into maintaining mainframe contracts. That reallocation, at scale across the Tier 1 peer group, has direct read-across to the enterprise software market, to cloud infrastructure capex forecasts, and to the valuations of legacy IT service providers whose managed-mainframe revenues depend on migration inertia persisting. Thought Machine, which has raised more than £500 million in funding and works with over 65 financial institutions globally, is making a direct bet that the inertia has finally run out.