The AI workforce reality
Across higher education, generative AI augments academic and research roles — accelerating drafting, literature work and assessment design — while the substitution pressure falls on student services, admissions processing, records, HR and finance operations. The split mirrors the IMF’s exposure-versus-augmentation distinction (Cazzaniga et al. 2024).
Many higher-education institutions are public authorities for the purposes of the Public Sector Equality Duty, which means workforce restructuring carries the same Equality Impact Assessment expectations as core public bodies, alongside UK GDPR obligations over staff and student data.
Redeployment pathways are broad: exposed professional-services staff carry transferable analytical and coordination skills that map to data, compliance and people-operations roles, and the DfE Unit for Future Skills shortage index points to durable demand in several of these destinations.
Where the exposure sits
| Role family | Type | Horizon | Exposure |
|---|---|---|---|
| Clerical & Administrative | high substitution | 1-3 years | 84 |
| Customer Operations | high substitution | 1-3 years | 78 |
| Junior Analysts (Generalist) | augmentation | 2-5 years | 64 |
| HR Operations | medium substitution | 2-5 years | 58 |
| Senior Professional / Knowledge Worker | augmentation | 3-7 years | 38 |
Where the workforce can go
The duties that bear on this sector
Public Sector Equality Duty (Equality Act s.149)
Public sector clients — central government, devolved government, local authorities, arms-length bodies, NHS trusts.
UK GDPR & Data Protection Act 2018 (as updated by DUAA 2025)
All workforce data processing — universal. Data Use and Access Act 2025 phased in during 2026.
Equality Act 2010
All redeployment, restructuring, and selection decisions — universal.
ICO Guidance on Automated Decision-Making and Profiling
Any AI-assisted scoring or recommendation in workforce decisions with significant effect on individuals.
Full requirements for each duty are set out on the regulatory landscape map. Forta scores roles, not individuals — the audit itself does not trigger automated-decision provisions.