In early December, I shared news about Google’s experiment with AI-generated headlines for articles. Initially, it seemed like Google was backing away from it, but now they’ve confirmed to The Verge that these AI headlines in Google Discover are here to stay. They claim this feature enhances user satisfaction, but many users, including myself, still find it misleading.
Think of it like a bookstore swapping out book covers—with Google acting as a digital news distributor. You swipe right on your phone to read the news, but what you see might not always tell the true story. For instance, last week, Google AI stated that the “US reverses foreign drone ban,” linking to a PCMag article that explicitly said the opposite. PCMag clarified that while the Commerce Department decided to end its efforts to restrict drone imports, they did so because those restrictions were already redundant.
Jim Fisher from PCMag expressed his frustration, stating, “I’d encourage people to click on stories and read them, and not trust what Google is spoon-feeding them.” He believes Google should rely on actual headlines written by humans instead of relying solely on AI.
Google insists they aren’t rewriting headlines but showcasing what they call “trending topics.” However, these trending topics look very much like our original articles and often lack proper fact-checking.
Interestingly, there have been some improvements. I’ve noticed fewer blatant clickbait headlines. Now, instead of artificially short titles like “AI tag debate heats,” the AI generates slightly longer, yet still confusing, headlines. But there’s still a huge problem: the AI doesn’t grasp the story’s significance or accuracy, often mixing stories together.
On January 20th, Google claimed, “3D tech wows,” linking to a story about different companies altogether. This confusion can mislead readers, who may not understand the news context.
Moreover, Google seems not to address the worst human-generated clickbait. A headline about a limited-time giveaway for a Star Wars game, for example, misled readers into thinking many copies were available when, in fact, only one was.
A Google spokesperson shared that these AI-generated overviews are designed to help people explore topics across various sources, but the effectiveness of this feature remains in question. It’s worth noting that The Verge’s parent company, Vox Media, has even filed a lawsuit against Google over its ad practices.
As we navigate through this changing landscape, it’s essential to question what we read. Relying solely on AI-generated summaries can lead to misinformation. So, always tap into the original stories for the full picture.
Source link
Google,Tech,Web
