PwC says 8% of firms have monetised AI. What is everyone else waiting for?

Two numbers that should not sit comfortably

PwC just published its 2026 Law Firm Survey. Two contrasting numbers within it should make you pause. Two-thirds of the UK Top 100 (66%) have now implemented AI, up from 43% a year ago. Only 8% report that they have monetised the benefit, down from 15% last year.

PwC's own assessment is blunt: deployment "continues to outpace the realisation of value". Expectations have moved the wrong way too. 78% of firms now expect AI to have an adverse commercial effect, up from 66% last year and 31% in 2023.

Why the benefit is hard to see

There is significant variation in firms' claims and a few report they are seeing very good results. Yet on average firms report saving less than 5% of chargeable hours. This is set against their own expectation of savings that average 18%. That is a large gap, but it is not the most telling number in the survey. Only 5% of firms have well-established ways of measuring the benefit at all.

A PwC presenter described the pattern as "broad but very shallow", and said firms are struggling to monitor the tangible commercial benefits. Many report real gains in quality and responsiveness to clients. Those gains are genuine. They are also very hard to turn into a figure a partner or a client will accept.

The call every firm will get

Clients read the same headlines. Sooner or later one of yours will ring and say that, with all this AI, they would like to see the fees come down.

What can you say? Do you know what the work cost to deliver before the tool arrived, and what it costs now? Do you know which matters the time came out of, and who kept the saving? Could you show the client where value has improved, or agree a price that shares the gain on terms you have worked out in advance?

If the answer is that you would need to go and find out, the benefit is not yet being realised, whatever the adoption figure says. The survey does not explain why monetised benefit fell. It may be that early claims met scrutiny. It may be something else. It is a good question to put to your own leadership team.

What is lagging is not the technology

PwC points to where firms are falling behind: pricing, incentives and the wider operating model. Only 23% have well-established training, and 13% have well-established data quality and access. Governance is further ahead, with 70% reporting well-established AI policies. PwC also notes that the survey does not yet show AI-enabled firms outperforming on growth or margin.

Most firms do not have a technology problem. They have a value realisation problem. A tool saves an hour, while implementing it introduces new cost. Whether that hour becomes margin depends on how the matter is priced, how fee earners are measured, who is accountable for the result, and whether anyone has defined what a good outcome would look like before the tool was bought.

Where Morphis comes in

Morphis does not sell or implement AI. We work on the operating model the tools land in: governance, process, people, technology, and data and reporting.

For a firm that has already spent the budget, the Diagnostic starts from your own numbers and shows where the value went. It packs a proven comprehensive approach into a rapid six week exercise that gives you clear priorities to recover value. If it finds nothing worth acting on, or we are not the right fit, we will say so.