Are your staff actually using AI properly?
Mortgage lenders and brokers across Britain convinced they are making strides on artificial intelligence may be looking at the wrong numbers, according to a new report that should give senior leaders pause.
A survey of 5,000 knowledge workers at large firms in the US, UK and Canada finds that three years after ChatGPT's launch, only 2.7% of employees qualify as "AI practitioners" – people who have woven AI into their workflows and are achieving real productivity gains. Just 0.08% meet the bar for "AI experts". In total, 97% of the workforce is using AI poorly or not at all.
For UK mortgage firms that have spent the past year rolling out AI tools, drafting policies and running training sessions, the findings are a sharp warning: most employees still do not know what to do with AI in the context of their actual jobs.
From prompts to productivity
The report argues that "AI proficiency" meant one thing in 2025 and something harder in 2026. Last year's focus was on basic literacy: understanding what AI is, how to avoid data leaks and how to write a decent prompt. Many firms can now tick those boxes. Staff know how to ask AI to tidy up an email to a broker or summarise a valuation report.
But the bar has shifted. Proficiency now means incorporating AI into meaningful, value-adding tasks every week – not sporadic experiments. That might be automated extraction of income data from bank statements, first-pass affordability checks, risk flagging in complex cases, or systematic review of compliance documentation.
The survey suggests most organisations have not come close. Seventy percent (70%) of workers are "AI experimenters": they use AI for very basic tasks such as summarising notes, rewriting emails and getting quick answers. Another 28% are "AI novices" who rarely or never use AI. Less than 3% are practitioners or experts.
The impact on time saved is predictably thin. A quarter of workers report saving no time at all with AI; another 44% save less than four hours per week. Only 6% say they save more than 12 hours weekly – the level that might begin to show up in processing costs and turnaround times.
The 'use case desert'
The central problem is not that people cannot prompt. It is that they do not know what to use AI for in their specific role.
Across respondents, 26% say they have no work-related AI use case at all; 60% say their use cases are beginner-level. After reviewing 4,500 reported work use cases, researchers judged only 15% likely to deliver real ROI.
The most common "most valuable" use case is using AI as a Google search replacement (14.1%), followed by draft generation (9.6%) and grammar editing (5.7%). Task and process automation – the category mortgage firms ought to care most about – sits at just 1.6%.
Overall, 59% of use cases are basic task assistance, more than a quarter have no role in larger workflows, and only 2% are rated advanced.
For a mortgage lender or broker, that often means staff ask AI to polish a customer email but do not use it to systematically extract data from payslips, pre-populate fact-finds, flag affordability edge cases or generate first-pass product recommendations. AI remains a clever spell-checker rather than a re-engineered origination or underwriting engine.
Executives in the dark
The most uncomfortable finding is the gap between what leaders think is happening and what employees experience.
Among C-suite respondents, 81% say their company has "a clear, actionable policy that effectively guides AI use". Only 28% of individual contributors agree – a 53-point gap. Eighty per cent (80%) of executives say tools exist with clear access; just 39% of frontline staff concur.
On access and support, the skew is sharp. Only 32% of individual contributors report clear access to AI tools, versus 80% of the C-suite. Just 27% of ICs have received company AI training, compared with 81% of executives. Only 7% of ICs are reimbursed for AI tools, versus 63% of senior leaders.
For UK mortgage firms, where much of the repetitive, rules-based work – income verification, document chasing, case preparation, compliance checks – is done by individual contributors, that skew is costly. The people whose work could most benefit from AI are the last to get tools, training and encouragement.
Training is not enough
Firms are investing. Sixty-three percent (63%) of respondents say their company has an AI policy, 50% have access to a tool and 44% receive training. These efforts do move the needle: employees at firms with a strategy are 1.6 times more proficient than those without; those whose managers expect AI use are 2.6 times more proficient.
Yet after all that, the average trained worker scores 40 out of 100 in proficiency. Most remain in the "experimenter" band.
The explanation is straightforward: training is still aimed at the wrong target. It teaches access, safety and prompting – how to use AI – but not how to identify and redesign workflows where AI can actually remove work. Knowing that AI can summarise a document is not the same as knowing how to redesign the mortgage application journey around it.
What UK mortgage leaders should do
For mortgage executives, the report implies several shifts:
-
Stop measuring success by logins and course completions. If only 15% of use cases are value-driving, adoption metrics are misleading. Track time saved per process and business outcomes instead.
- Treat use case development as a managed competency. Build function-specific libraries for underwriting, processing, compliance and broker support, and make use case development a formal objective for team leaders.
- Prioritise individual contributors. Standardise access, ensure fair reimbursement and require managers to identify at least three AI use cases for each direct report.
- Shift training from "how to prompt" to "how to redesign workflows". Teach teams to map processes, spot bottlenecks and test AI in controlled slices of work.
- Close the executive awareness gap. Shadow staff as they attempt to use AI in daily work and insist on hard measures such as hours saved and error rates, not just dashboard stats.
The technology is ready. The question is whether UK mortgage firms can redesign how work is done quickly enough to turn that capability into lower processing costs, faster decisions and genuinely better customer outcomes – rather than another layer of digital polish on unchanged processes.
