Google is currently trialing an AI-powered feature within its Discover news feed that automatically rewrites editorial headlines—often shortening them, making them more sensational, or even altering their meaning inaccurately. The outcome contradicts Google’s own Discover guidelines, which explicitly discourage engagement-driven clickbait titles.

Some of these AI-generated rewrites have drawn attention for their inaccuracy. For instance, a headline from Ars Technica originally reading “Valve’s Steam Machine Looks Like a Console, But Don’t Expect It to Be Priced Like One” was replaced with the misleading “Steam Machine Price Revealed.” Similarly, a detailed report titled “Radeon RX 9070 XT Outsells Entire NVIDIA RTX 50 Series at Major German Retailer Mindfactory” was oversimplified to “AMD GPU Surpasses Nvidia.”

Google stated that this feature is part of a small-scale experiment designed to help users grasp article content more quickly. However, by substituting newsrooms’ editorial judgment with its own automated system, the company is raising concerns about its growing influence over how news is framed and delivered—extending beyond its already significant role in content distribution.