Did you know that just 1% of users who see Google’s new AI summaries actually click on links to other websites? That’s right, this daunting statistic reveals a shocking reality in the digital world!

Yesterday, the Pew Research Center unveiled a striking report based on the online behavior of 900 adults in the U.S. The findings show that users encountering AI summaries provided by Google are significantly less likely to explore additional sites compared to those who don’t see these summaries. In fact, only 1 percent of users who come across an AI summary venture to the actual site that Google is summarizing!

This shift illustrates a dramatic change in how Google, which has long been the internet’s traffic powerhouse, is operating since it introduced its AI Overview feature in 2023. This new format is poised to drastically cut down the traffic that flows to countless blogs and news sites, effectively putting many of them at risk of extinction. What’s even more unsettling is that this AI tool often presents information with such unwarranted confidence that users might not recognize the errors it contains.

Let’s take a moment to understand the implications of this. From the perspective of someone dedicated to crafting and sharing valuable information online, I recently published an article detailing Spotify's decision to release AI-generated songs from deceased artists—without their permission! After a full day of research and outreach, I managed to spotlight this issue, prompting Spotify to remove the offending tracks. This all reflects the delicate balance of trust that must exist between creators and their audiences.

My Spotify article attracted a fair amount of traffic, with Google being the leading source. However, it’s frustrating to realize that it could have sent even more readers my way. When I searched for 'AI music Spotify' on Google, the first result was a summarization of my article... not from my site, but from a third-party blog that replicated my work. Instead of directing traffic to me, Google linked to several other sites that had pooled together information, robbing me of the opportunity to engage directly with readers.

This is a widespread problem. Every website, not just mine, is facing the possibility of being overlooked in favor of aggregated content that offers little value. The unsettling consequence of this “traffic apocalypse” could decimate small businesses and independent creators, undermining the very foundation of the internet where original thought and creative expression should thrive.

To make matters even stranger, let’s not forget some of the bizarre errors that have surfaced from these AI summaries. Remember the time Google’s AI suggested users “eat glue”? This gem came from a misunderstanding of a humorous Reddit conversation, but it highlights how seriously flawed this technology can be. Disturbingly, it’s not just trivial mistakes; it’s also about the potential for misinformation to be propagated. Take artist Eduardo Valdés-Hevia, who jokingly invented fictional theories only to discover that Google’s AI was presenting them as factual! This raises a chilling question—how easy is it for misinformation to seep into the mainstream?

As we grapple with these technological advancements, we must question whether Google, having dominated search for over two decades, is becoming vulnerable to disruption. There’s a growing call for human-centered alternatives to AI-powered search engines, a shift that could restore sanity to the information landscape. However, if we continue down this path, we may be paving the way to an internet where distinguishing truth from fiction becomes increasingly challenging. And that, my friends, is a consequence we cannot afford to ignore.