Google BAFTA automated news alert includes "N-word"
Google’s automated news system broadcast a racial slur to users' devices by incorrectly "correcting" a euphemism into the full slur, causing significant racial trauma and highlighting the failure of safety filters to prevent the amplification of hate speech. What happened John Davidson, an activist with Tourette syndrome, involuntarily shouted the N-word during a presentation at t he BAFTA Film Awards by Black actors Delroy Lindo and Michael B. Jordan. While the incident itself caused significant distress, it was worsened when a Google News automated notification system sent a push alert inviting users to "See more on [N-word]." Many recipients and observers described as shocking and offensive, especially during Black History Month in the US. Google later removed the notification and issued an apology, claiming that only a limited number of users saw it. Why it happened Google later said the problem stemmed from an error with an automated system it uses to recognise and provide "clarity" on euphemisms for offensive terms appearing on web pages. According to the tech company, the filters meant to catch and block the use of slurs did not trigger and the system incorrectly "applied" the full offensive term to the notification text after identifying a euphemism in news articles it was linking to. The incident reflects indequate human-in-the-loop oversight of automated push notification systems, especially for high-stakes, sensitive news topics. What it means For those impacted , the incident reinforces how automated systems can inappropriately reproduce and amplify racist language, placing the burden of harm on Black users and other affected communities. For society and policymakers , it underlines the need for clearer expectations around “safety by design” for algorithmic news curation systems, including testing for edge cases involving hate speech, slurs, and protected characteristics. System 🤖 Unknown Developer: Google Country: UK Sector: Media/entertainment/sports/arts Purpose: Recognise and clarify euphemisms Technology: NLP/text analysis Issue: Accuracy/reliablity; Safety
- Date it happened
- 2026-02-01
This incident was imported from AIAAIC and is used under CC BY-SA 4.0. Our additions to it — the structured fields, the translation, the checks against other reports — are published under the same licence.
This is a record of what was reported, not a finding that anyone broke the law. If it names your organisation and you believe it is wrong, the corrections process is free and open to everyone.