Some goofy results from ‘AI Overviews’ in Google Search. John Gruber collects two of the best examples of Google’s new AI overviews going horribly wrong.
Gullibility is a fundamental trait of all LLMs, and Google’s new feature apparently doesn’t know not to parrot ideas it picked up from articles in the Onion, or jokes from Reddit.
I’ve heard that LLM providers internally talk about “screenshot attacks”—bugs where the biggest risk is that someone will take an embarrassing screenshot.
In Google search’s case this class of bug feels like a significant reputational threat.
Recent articles
- The last six months in LLMs, illustrated by pelicans on bicycles - 6th June 2025
- Tips on prompting ChatGPT for UK technology secretary Peter Kyle - 3rd June 2025
- How often do LLMs snitch? Recreating Theo's SnitchBench with LLM - 31st May 2025