Technology news and discussion on TalkFuse.
Computer says d’oh: Hours at work are spent fixing AI mistakes
UK employees are spending nearly seven hours a week correcting AI-generated content, including removing unnecessary wording and fixing inaccurate facts and figures. The findings suggest that while AI is widely used and often saves time, unchecked errors can shift work to colleagues and make its overall effect on productivity less clear.
A survey of 2,000 UK workers commissioned by Arden University found that 64 per cent had used AI at work in the previous three months, mainly for research, emails and ideas; 84 per cent of AI users said it saved them time overall. Nineteen per cent had completely redone a colleague’s AI-generated work, 53 per cent did not tell management when they used AI, and only 13 per cent felt confident spotting fake sources. Arden says better training could help staff give clearer instructions, check accuracy and use AI responsibly.
In short
- Workers spend nearly seven hours a week fixing AI-generated content.
- 84 per cent of AI users say it saves them time overall.
- Arden University says better training could improve responsible use.
Steelman: both sides, in good faith
The strongest fair case each way — we don't pick a winner.
The case for
AI can save employees time on routine research, drafting and idea generation, and most surveyed users said it did so overall. With suitable training and careful checking, workplaces may retain those gains while reducing errors, freeing staff for work that needs human judgement.
The case against
The time saved for one employee can become extra work for colleagues if AI output contains inaccuracies, padding or unreliable sources. Widespread use without disclosure or confidence in checking results makes those costs harder to track, so organisations may struggle to know whether AI is improving productivity overall.