Deloitte UK’s reported plans to significantly expand its AI assurance practice could have important implications for software testing and quality engineering teams across banking and financial services.
According to reports, the professional services firm is moving increasing numbers of auditors towards AI assurance work as demand grows for services focused on AI governance, model validation, controls testing and regulatory compliance.
While the move may initially appear to be an audit story, it highlights a broader shift taking place across the financial services industry: the emergence of AI assurance as a dedicated discipline focused on independently testing, validating and monitoring increasingly complex AI systems.
The development comes as banks accelerate their adoption of generative AI, autonomous software development tools and AI-powered testing platforms. As a result, organisations are increasingly grappling with a fundamental question: who validates the AI?
Historically, software testing teams have focused on validating functionality, performance, security, resilience and regulatory requirements.
However, AI systems introduce entirely new categories of risk, including hallucinations, model drift, bias, explainability, governance failures and unpredictable agent behaviour.
As financial institutions deploy AI into customer service, fraud detection, software development, compliance monitoring and operational processes, the challenge of assuring those systems is rapidly becoming a testing issue as much as an audit or governance one.
Deal with UiPath
The trend is already visible in Deloitte’s own technology and quality engineering work.
Earlier this year, Deloitte expanded its collaboration with UiPath around agentic AI-driven software testing through the Deloitte Ascend delivery platform.
The joint initiative combines UiPath Test Cloud with Deloitte Ascend to automate test design, reduce maintenance effort and introduce self-healing testing capabilities.
According to the companies, the platform enables “continuous autonomous quality” and can “proactively detect changes, generate tests, and execute them autonomously.”
Deloitte and UiPath said the platform incorporates more than 1,500 prebuilt testing bots and domain-specific AI agents designed to help organisations automate increasingly complex testing activities.
The announcement reflects a wider transformation taking place across software engineering and quality assurance.
“We’re entering a new era in software engineering, and AI is at the centre of it,” Ryan Lockard, Engineering Principal and Banking & Capital Markets Lead at Deloitte, wrote in a recent analysis examining the impact of AI on banking software development and testing.
“From test case generation to continuous integration, AI is fundamentally transforming the way banks ensure software quality.”
According to Lockard, many financial institutions are already using AI to automate significant parts of the testing lifecycle.
“We’re seeing banks adopt AI tools that can automate significant parts of the testing process,” he explained.
“Solutions like Applitools, Functionize, and Testim are already being used to generate test cases automatically, execute thousands of tests rapidly, and even learn from previous runs to improve outcomes over time.”
That evolution is important because AI assurance increasingly requires capabilities that traditional software testing teams already possess. Independent validation, controls testing, evidence gathering, risk assessment, monitoring and continuous verification have long been core competencies within quality engineering organisations.
The difference is that the systems being tested are changing. “Natural language processing is now being used to convert business requirements into test scenarios, catching inconsistencies and gaps early,” Lockard explained. “This is especially valuable in banking, where minor errors can have major consequences.”
“We’re seeing banks adopt AI tools that can automate significant parts of the testing proces.”
– Ryan Lockard
The emergence of AI assurance suggests that testing professionals may increasingly find themselves validating not only software applications but also the behaviour, outputs and governance controls surrounding AI models and autonomous agents.
For banks, this challenge is becoming particularly acute as AI adoption accelerates.
Citizens Bank recently reported productivity gains from pilots involving generative AI in development and testing workflows, while Citigroup has equipped tens of thousands of developers with generative AI tools to support software modernisation efforts.
At the same time, institutions continue to modernise legacy systems, migrate critical workloads and deploy AI-powered services into production environments.
According to Lockard, those transitions cannot succeed without robust validation processes.
“A significant portion of the banking core still runs on COBOL,” he stressed. “AI models like IBM’s watsonx Code Assistant are helping teams convert this old code to Java, but the transition can’t succeed without reliable, AI-driven testing frameworks to validate every step.”
Traditional processes
The growing demand for AI assurance services reflects a recognition that traditional governance processes alone may not be sufficient.
“AI is uniquely suited to tackle these inefficiencies,” Lockard explained. “Banks can deploy AI to optimise legacy testing frameworks, improve test data management, and accelerate feedback loops. This results in fewer bugs, faster remediation, and lower long-term maintenance costs.”
At the same time, Deloitte has repeatedly argued that AI adoption must be accompanied by strong controls, governance and oversight.
“Implementing AI in testing doesn’t eliminate risk, it changes it,” Lockard said. “You need clear policies, data controls, and oversight mechanisms to manage things like false positives, security testing, and compliance.”
“Banks deploy AI to optimise legacy testing frameworks, improve test data management, and accelerate feedback loops. Fewer bugs, faster remediation, and lower long-term maintenance costs.”
– Ryan Lockard
For QA teams, Deloitte’s AI assurance expansion may offer an early indication of where the profession is heading.
The software testing industry has already evolved from manual testing to automation, from automation to continuous testing and increasingly towards autonomous quality engineering.
The rise of AI assurance represents another step in that evolution, one that combines traditional testing disciplines with governance, risk management and AI validation expertise.
As financial institutions continue deploying AI across critical systems and business processes, demand is likely to grow for professionals capable of independently validating not only whether software works, but whether increasingly autonomous AI systems are behaving as intended.
“Ultimately,” Lockard concluded, “this is about building trust through technology. The more we can automate, standardise, and accelerate software testing, the better we can serve customers, regulators, and the market at large.”
16 SEPTEMBER IN LONDON
REGISTER TODAY – SIMPLY CLICK HERE
Why not become a QA Financial subscriber?
It’s entirely FREE
* Receive our weekly newsletter every Wednesday * Get priority invitations to our Forum events *
REGULATION & COMPLIANCE
Looking for more news on regulations and compliance requirements driving developments in software quality engineering at financial firms? Visit our dedicated Regulation & Compliance page here.
READ MORE
- Inside Rabobank: Engineering resilience by design
- Can AI agents finally automate data testing?
- Continuous testing drives DORA compliance
- Why software testing may face a major rethink
- Buy or build? AI rewrites software testing for banks
WATCH NOW
QA FINANCIAL PODCASTS
CLICK HERE TO LISTEN TO OUR EXCLUSIVE CONVERSATIONS
