Google Is Finally Trying to End AI Clickbait


Google is taking action against algorithmically generated spam. The search engine giant recently announced upcoming changes, including a revised spam policy, partly designed to keep AI clickbait out of its search results.

“It looks like this is going to be one of the biggest updates in the history of Google,” says Lily Ray, senior director of SEO at marketing agency Aimsive. “This could change everything.”

In a blog post, Google claims the change will reduce “low-quality, unrealistic content” in search results by 40 percent. It will focus on reducing what the company calls “scaled content abuse”, which occurs when bad actors flood the Internet with massive amounts of articles and blog posts designed to game search engines.

“A good example of this, which has been around for a while, is the abuse around obituary spam,” says Pandu Nayak, Google's vice president of search. Obituary spam is a particularly serious type of digital theft, where people attempt to make money by scraping and republishing death notices, sometimes on social platforms such as YouTube. Recently, obituary spammers have begun to use artificial intelligence tools to enhance their output, making the problem even worse. If Google's new policy is implemented effectively, it will become harder for this type of spam to appear in online searches.

This particularly more aggressive approach to combating search spam takes specific aim at “domain squatting”, a practice in which scavengers purchase websites with name recognition to profit from their reputation, often manipulating search engine rankings. Replace original journalism with AI-generated articles designed to. This type of behavior predates the AI ​​boom, but with the rise of text-generation tools like ChatGPT, it's become easier to churn out endless articles to game Google rankings.

The rise in domain squatting is one of the issues that has plagued the reputation of Google Search in recent years. “People can open these sites really easily,” says SEO expert Gareth Boyd, who runs digital marketing firm Fort Analytica. “It's been a big issue.” (Boyd admits that he has created similar sites in the past, although he says he no longer does so.)

In February, WIRED reported on several AI clickbait networks that used domain squatting as a tactic, including the websites of shuttered indie women's website The Hairpin and shuttered Hong Kong-based pro-democracy tabloid Apple Daily And they were filled with AI. -Generated nonsense. Another turned a small-town Iowa newspaper's website into a quirky repository for AI blog posts on retail stocks. According to Google's new policy, this type of behavior is now explicitly classified as spam by the company.

In addition to domain squatting, Google's new policy will also focus on eliminating “reputation abuse,” where otherwise trustworthy websites allow third-party sources to publish junk sponsored content or other digital junk. (Google's blog post describes “Payday loan reviews on trusted educational website” as an example.) While other parts of the spam policy will begin to be implemented immediately, Google will allow websites to crack down on reputation abuse. Giving 60 days notice before. Queuing time.

Nayak says that the company has been working on this specific update since late last year. More broadly, the company is working on ways to fix low-quality content in Search, including AI-generated spam, from 2022. “We are aware of the problem,” says Nayak. “These changes take time to develop effectively.”

Some SEO experts are cautiously optimistic that these changes could restore Google's search efficacy. “Hopefully things will go back to the way they were,” says Ray. “But we'll have to see what happens.”