- Published on
Don’t be fooled. The AI that was supposed to save us time is doing just the opposite | Gaby Hinsliff
- Authors
- Name
- DeepHorus
Artificial intelligence is creating new administrative burdens rather than relieving them, according to Gaby Hinsliff’s latest column. She argues that tools marketed as time-savers are generating floods of machine-written correspondence, complaints and applications. Humans must then spend time sorting and disentangling the output.
Hinsliff compares the situation to email, which promised to reduce meetings and calls but instead produced endlessly refilling inboxes. Her own inbox holds more than 7,000 messages even after a cull. The ease of reaching the world in seconds, she writes, came with the world reaching back.
The trend threatens government hopes that AI can transform public services cheaply and save billions of pounds. Sam Freedman noted in a Substack post that governments have seen unusual surges in complaints and requests to official bodies since ChatGPT’s adoption. Employment tribunal cases in England, Scotland and Wales reportedly rose 39% in a year, partly because claimants use AI to draft claims.
Whitehall special advisers complain departments are drowning in AI-generated freedom of information requests. MPs’ offices are similarly swamped by AI-generated letters, which staff must read in case they contain genuine constituent concerns. Public consultations risk being spammed by machine-generated slop, undermining their value as a measure of public opinion.
Detection tools are not a simple fix, Hinsliff argues, especially when vulnerable people use AI to express themselves and still deserve human replies. Employers that used AI to screen job applications now face applicants automating their own applications and spraying them across the market. Neither side can separate wheat from chaff.
If agentic AI spreads widely, shortcuts could become bottlenecks, much as cars slowed to about 7mph in central London daytime traffic by 2018. The promise of frictionless administration may collapse under the weight of automated responses. Public systems could face paralysis unless human oversight and safeguards keep pace.
