Louise Haigh has discovered the solution to Britain’s trust crisis: replace politicians with machines. Speaking at the TUC Congress, she warned that citizens must heed AI safety concerns raised by tech developers—because apparently the people actively writing the code are more trustworthy than the people writing policy.
The irony is so thick you could legislate it. Haigh wants the public to take seriously the existential warnings from Silicon Valley while simultaneously ignoring the existential threat posed by her own colleagues, who have spent the last decade lying about immigration figures, tax compliance, and whether they attended parties during lockdown. An AI system trained on parliamentary records would simply refuse to function—the contradiction density would exceed its processing capacity.
Why should anyone trust warnings from tech developers over their elected representatives? Because tech developers have to live with the consequences of their code. Politicians, by contrast, have parliamentary privilege and a pension fund.
The speech positions AI safety as a matter of national urgency requiring immediate citizen cooperation. Meanwhile, actual human politicians continue operating without meaningful accountability, fact-checking, or even basic conflict-of-interest disclosures. One group is subject to congressional oversight and regulatory frameworks. The other is you and me. Guess which one Haigh wants us to trust.
The TUC was reportedly unmoved. They’ve seen the spreadsheets.