Topic
Tendency of AI and algorithmic systems to reproduce or amplify societal biases—such as gender or racial stereotypes—often traced to skewed training data.
Matt Wilson's AI agents represent close to half a million professionals to around 5,000 companies, and the bias experiment he cites as the reason to trust audited screening sits in his own company's white paper.
Reality
Since June 2026 the Western Australian Police Force scanned more than 900,000 faces, logged 209 alerts, made 79 arrests and recorded eight incorrect matches. What those eight mean depends on which total you divide them by.
Reality
Its contracts and third parties section asks AI buyers to decide and record controller, processor or joint controller roles across the whole supply chain, and it warns that the guidance is under review after the Data (Use and Access) Act.
Reality
A University of Washington test paired 554 real CVs with race- and gender-associated names against more than 500 job listings, and the models preferred white-associated names 85% of the time and female-associated names about 11%.
Reality
The Joint Committee on Human Rights says no country's rules on AI are fit for purpose, and it wants a statutory oversight body, tiered duties across the supply chain and some uses banned outright. It can only recommend.
Reality
Google's AI Mode began completing hotel bookings in the US in late August, with Booking.com, Expedia, Hilton, Marriott and IHG supplying rooms. The published evidence about what goes wrong is one chatbot case from 2022.
Reality
A structured test of ChatGPT, Claude and Perplexity found competent financial mechanics and a consistent blind spot for the user's circumstances. That argues for documented failure modes.
Reality