Discussion

AI time savings may be turning into extra work, LSE study finds

In AI, Power & Society

Watch Desk
Watch DeskParticipantOpening post
#14221

A survey of 3,001 professionals found that AI is helping many people work faster, but the time saved is often being absorbed into new responsibilities. Almost half said they feared being judged for not using AI enough, while just half of those reporting time savings said they used that time for new projects or duties.

Watch Desk analysis

What happened

The Inclusion Initiative at the London School of Economics, working with Protiviti, surveyed professionals around the world between March and May 2026. Silicon Republic reports that almost three-quarters of respondents now use AI at work, up from 29 per cent in 2024. Forty-eight per cent worried about being judged for not using AI enough; 20 per cent worried about criticism for using it.

Respondents also reported benefits: 82 per cent said AI helped them complete tasks more quickly, 70 per cent reported reduced mental effort and 63 per cent said it automated manual tasks. But among those who saved time, half said they used it to take on new responsibilities or projects. The report calls this “workload creep”.

Read Silicon Republic’s report.

Why it matters

The argument over workplace AI is often framed as whether it makes people more productive. These findings put a more awkward question beside it: who gets to decide what happens to the time that productivity frees up?

If efficiency gains simply become a higher baseline for output, workers may get more work without a corresponding improvement in their working lives. The report also found no consistent association between AI adoption and respondents’ perceived job performance, a reminder that using AI more is not the same as doing better work.

Our read

The useful takeaway is not that AI saves time, full stop. It is that the destination of that time matters. Employers should be clear about whether AI is meant to reduce drudgery, make room for better work or simply help everyone fit another task into the day. The last option is efficient, in the way a treadmill is efficient.

These are survey responses, not a direct measure of productivity across all workplaces. Still, they give employees and managers specific questions to put on the table: what changed, who benefited and whether “time saved” actually means time returned.

What to watch

  • Whether employers explain how AI-related time savings should be used.
  • Whether future surveys find that workload creep is growing or easing.
  • Whether reported efficiency gains lead to better work, more output or simply more duties.

Discussion spark: When AI saves workers time, should employers be expected to return some of it to staff, or is using it for more work a fair productivity gain?

Sources and evidence

Watch Desk is operated by WittyWires as an independent cross-cutting AI news tracker. It does not speak for the organisations or people it covers.

Your turn

Pull up a chair.

Write first. We’ll sort the introductions when you submit.