Platform

Solutions

Resources

Platform

Solutions

Resources

Platform

Solutions

Resources

Blog

AI adoption is accelerating. Is our ability to assure it keeping up?

Andrew Doughty

Published on

AI Testing
eBook CTA Background
RECOMMENDED RESOURCES

The Ultimate Enterprise Guide To Agentic Testing

AI agents can write, fix and triage your tests at scale. See how to use them in QA while keeping people in control of everything that ships.

Get the eBook →

Enterprise guide to agentic testing
eBook CTA Background
RECOMMENDED RESOURCES

The Ultimate Enterprise Guide To Agentic Testing

AI agents can write, fix and triage your tests at scale. See how to use them in QA while keeping people in control of everything that ships.

Get the eBook →

Enterprise guide to agentic testing
eBook CTA Background
RECOMMENDED RESOURCES
Enterprise guide to agentic testing

The Ultimate Enterprise Guide To Agentic Testing

AI agents can write, fix and triage your tests at scale. See how to use them in QA while keeping people in control of everything that ships.

Get the eBook →

eBook CTA Background
RECOMMENDED RESOURCES

The Ultimate Enterprise Guide To Agentic Testing

AI agents can write, fix and triage your tests at scale. See how to use them in QA while keeping people in control of everything that ships.

Get the eBook →

Enterprise guide to agentic testing

Table of contents

Lorem ipsum

Lorem ipsum

Lorem ipsum

Lorem ipsum

Lorem ipsum

The Bank of England and FCA published their latest report on AI in UK financial services this week. There are some important findings.

86% of responding firms are now using AI, up from 75% in 2024.

80% of firms responding to the detailed survey have AI agents deployed, with another 14% planning to deploy them over the next 12 months.

But the statistic that stood out to me was something else.

While more than 90% of respondents reported controls such as human approval and pre-deployment review, only around half reported ongoing monitoring, leakage testing and periodic review or recertification.

That is a significant gap.

And I think it points to one of the biggest challenges financial services organisations will face over the next few years.



AI Changes the Rate of Change

For years, software quality assurance has largely been organised around releases. Build something. Test it. Approve it. Deploy it. Then repeat.

That model was already under pressure from cloud applications and continuous delivery. AI accelerates the problem.

Software is being created and changed faster. AI capabilities are being embedded into existing applications and business processes. Third-party models and services introduce dependencies that organisations don't fully control.

The Bank's report makes that last point particularly clear: third-party implementations now account for 53% of reported high-materiality AI systems, compared with 35% in 2024.

An organisation may understand its own software, but can it continuously establish that the entire system including third-party components still behaves as intended?



A Successful Test isn't Necessarily Proof of a Successful Outcome

This is where I think testing needs to change.

Running more automated tests is useful. Generating them faster with AI is useful too. But neither necessarily answers the important question.

What evidence do we have that the system is doing what the business intended, and that the risks and controls we care about are working?

Consider an insurer using AI within claims processing.

It's not enough to prove that a claim can be submitted, assessed and paid. We need to establish that the correct policy terms were applied, the right approvals occurred, payment limits were respected, exceptions were handled appropriately and the outcome can be explained and evidenced. And we need to keep verifying those things as the application, business rules, integrations and AI components change.

That is a different problem from simply automating a workflow.



Quality Assurance Needs to Become Continuous

The Bank's findings reinforce something we have believed for some time at Virtuoso.

As the rate of software change increases, the rate at which we establish confidence in that software has to increase with it.

That means moving beyond static regression packs towards assurance that understands business objectives, application behaviour, risk, controls and the evidence required to verify an outcome.

It's why we built Touchstone.

Touchstone combines application knowledge with customer-specific requirements, configuration, business processes and risk context. Its agents use that knowledge to design, create, execute and maintain tests and assertions, with traceability back to the requirements and evidence of what actually happened.

The objective isn't simply to produce more tests. It's to make testing more effective, reduce the human effort required and give organisations better evidence on which to make decisions. And importantly, it doesn't remove the human from the process.

AI proposes. Humans govern.



Confidence Cannot Be Assumed

One other finding caught my attention. Only 2% of respondents reported deploying AI agents with broad autonomy.

That tells me financial institutions recognise the importance of control and accountability, even as adoption accelerates. But governance cannot just mean a person approving something before it goes live.

That person needs reliable, traceable evidence to make an informed decision. And the organisation needs to know that the evidence remains valid as things change.

The opportunity AI creates is enormous. So is the responsibility to make sure the systems we build and deploy continue to operate as intended.

AI is accelerating software change. Quality assurance cannot remain a periodic exercise.

We need to continuously prove what works, what doesn't and why. Because in financial services, confidence needs to be supported by evidence, not assumed.

If you operate in the Banking, Financial Services or Insurance domain and would like a free trial of Virtuoso QA Touchstone

Written by

Andrew Doughty

Chief Executive Officer

Andrew is CEO of Virtuoso QA. He writes about where software delivery is heading now that AI generates code faster than teams can verify it, and why release evidence, not test counts, is the metric that matters.

See what your next release looks like with Virtuoso

Book a walkthrough on your applications and your workflows. Bring a requirement, a user journey, or a brittle Selenium script, and watch the loop run on something you recognise.

See what your next release looks like with Virtuoso

Book a walkthrough on your applications and your workflows. Bring a requirement, a user journey, or a brittle Selenium script, and watch the loop run on something you recognise.

See what your next release looks like with Virtuoso

Book a walkthrough on your applications and your workflows. Bring a requirement, a user journey, or a brittle Selenium script, and watch the loop run on something you recognise.

Virtuoso QA is establishing the standard of proof for software releases. Its governed QA loop turns business requirements into tests for any browser-based application: AI proposes, a deterministic engine executes, a person approves what matters, and every decision leaves evidence.

Trust Center

AICPA

SOC

WAVE STRONG PERFORMER

@ Copyright 2026 SpotQA, Creators of Virtuoso QA

Virtuoso QA is establishing the standard of proof for software releases. Its governed QA loop turns business requirements into tests for any browser-based application: AI proposes, a deterministic engine executes, a person approves what matters, and every decision leaves evidence.

Trust Center

AICPA

SOC

WAVE STRONG PERFORMER

@ Copyright 2026 SpotQA, Creators of Virtuoso QA

Virtuoso QA is establishing the standard of proof for software releases. Its governed QA loop turns business requirements into tests for any browser-based application: AI proposes, a deterministic engine executes, a person approves what matters, and every decision leaves evidence.

Trust Center

AICPA

SOC

WAVE STRONG PERFORMER

@ Copyright 2026 SpotQA, Creators of Virtuoso QA