AWA Research · May 2026
AI Impact Report 2026
Across 412 UK occupations, AWA’s May 2026 model finds AI gives the average worker 10.9 hours back a week — 15.8M FTE of capacity, and a serious set of questions about workforce, office space and burnout. Three decades of workplace science, one task-level model.
Central scenario = medium displacement assumption applied over a 5-year horizon. The model also offers Low, Medium and High scenarios across 3, 5 and 10-year horizons.
A summary in three effects
Substitution. Amplification. Cognitive relief.
Three measurable effects of AI on UK work — each calling for a different organisational response.
AI absorbs existing work — admin, data processing, customer service. Adds up to 10.1M FTE of capacity across the UK.
AI enables work that wasn’t previously feasible — broader analysis, deeper client service. A further 5.7M FTE on top of substitution.
In healthcare, education and the public sector, AI carries the admin load. Less burnout, longer careers, more time for the core work.
What’s in the report
Since 1992 AWA has worked with Microsoft, Google DeepMind, X (formerly Twitter), and organisations across banking, healthcare and government — applying neuroscience and behavioural science to how work actually gets done.
The 412-occupation model is the next step in that work, built so leaders can act from evidence, not speculation. Here’s what’s inside:
412 occupations, task-level
A decomposition of every UK occupation into 18 task categories, each scored 0–1 for AI susceptibility. Three distinct effects modelled: task substitution, capability amplification, and cognitive relief. Scenario projections across 3, 5, and 10-year horizons.
CEO, CPO, CRE Director
Written for the leadership team responsible for turning AI capability into organisational outcomes — from workforce planning and retention to real estate strategy and burnout response.
Evidence over speculation
Methodology, the full 412-occupation breakdown, sector-specific implications, and the framework for translating national figures into projections for your own organisation.
AI’s third effect: cognitive relief
For most of the AI conversation so far, two effects have done the heavy lifting — displacement and productivity. The May 2026 update to AWA’s analysis names a third that matters at least as much, particularly in the sectors where burnout is most acute.
In healthcare, education, financial services and the public sector, freed time is most usefully framed as recovery. The time and mental bandwidth professionals need to do their core work well, and to sustain their careers without burning out.
“For years, organisations have measured workload almost entirely through time, hours worked, meetings attended and tasks completed. But the real issue in many modern workplaces is cognitive overload.”
— Andrew Mawson, Founder, AWA
Hours freed per worker per week
Task substitution and capability amplification, stacked by occupation group
Source: AWA AI Impact Report 2026, Figure 2. UK-wide average: 10.9h substitution + 6.2h amplification = 17.1h combined.
What the research found
Practical AI impact follows a clear three-tier pattern across all 412 UK occupations.
Admin, data processing, customer service, teaching
The highest practical impact sits here. The right organisational response should differ by sector: displacement for back-office processing, cognitive relief for high-burnout teaching and care.
Solicitors, consultants, engineers, researchers
The dominant effect here is amplification. Freed time is best redeployed into deeper analysis, broader client service and more strategic work.
Trades, drivers, carers
Current AI offers limited gains for embodied work. The administrative components of these roles — care documentation, compliance paperwork, scheduling — can still see meaningful relief.
Capacity is the easy part. The harder question is what you do with it.
The model identifies three distinct effects, and which one dominates depends on the role, the sector and the organisation’s strategic intent.
For office-based organisations, the central scenario produces a reduction in effective office demand that exceeds 11% when combined with continuing hybrid working. For those already running desk-share ratios, the arithmetic for portfolio rationalisation is compelling, particularly where leases approach break clauses.
The broader point is harder to quantify. In sectors where 91% of UK adults report high or extreme pressure and one in five take time off for stress-related mental health, AI’s capacity to absorb administrative burden goes well beyond a productivity story. It moves into the territory of duty of care.
“Even small reductions in mental workload can have a disproportionately positive impact on wellbeing, focus and performance.”
— Andrew Mawson, Founder, AWA
Projected UK office jobs displaced
Thousands of jobs, by scenario and time horizon
Source: AWA AI Impact Report 2026, Figure 1 (displacement scenarios). Central estimate (medium, 5-year): 929,000 jobs displaced; 88M sq ft office space reduction.
Full report
Get the full analysis
10 sections covering methodology, the 412-occupation breakdown, displacement scenarios, sector implications, and the three-effect framework — substitution, amplification, cognitive relief.
About AWA
AWA is the leading global independent consultancy transforming the world of work, using science, research and depth of experience to deliver exceptional results for exceptional companies. Since 1992, we’ve worked at the intersection of evidence and practice — the DNA of work.
