The UK’s Copilot experiment with 20,000 civil servants deserves way more attention than it’s gotten. The results, 26 minutes saved per day, might seem modest, but they reveal something crucial about AI in government.
When you put this alongside the Alan Turing Institute’s finding that 41% of public sector work could theoretically be automated, you get a telling gap: 41% potential vs. 5% realized gains. That’s not a technology problem, it’s an institutional one.
I took some time to dig into what this really tells us about GenAI in government: why most deployments stay stuck at light-touch augmentation, what the political economy of digital transformation actually looks like, and why the path from pilots to systemic change requires confronting some uncomfortable truths about how government actually works.