The Weird, Yet Useful Trick That Seems To Turn Off Google AI Search Results

Google has been progressively integrating artificial intelligence into search for years, but this has escalated with the recent advent of generative AI. As of early 2025, features like AI Overviews of Google search results are now available in well over a 100 countries. Instead of being presented with a traditional page of search results, users now receive AI-generated summaries that aggregate information from various sources. According to Google, AI-powered search results work particularly well when you "want to quickly understand information from a range of sources, including information from across the web." 

While probably useful in some situations, AI summaries can be rather annoying when their reliability can be questionable, especially when it occupies half the screen space with each search. What good does it even do to scroll through potentially inaccurate AI-generated text when you want to read real perspectives from other human beings, shaped by their lived experiences? No wonder everyone is appending their searches with "reddit" these days.

This would be a non-issue if Google made AI Overviews and similar features optional, but they are on by default with no way to turn them off. However, there is one strange trick that still works.

As of January 2025, the easiest way to turn off AI search results is to add profanity to your search queries, as strange as that may sound. This trick has apparently been circulating on Tumblr, but we first learned about it through Bluesky user Nino Cipri. Let's say you want to find out how to stop your dog from snoring as much. If you search "how do I stop my dog from snoring," you'll encounter an AI-generated overview on Google. But if you search "how do I stop my f**king dog from snoring" instead, you'll get standard Google search results, like in the pre-AI days.

This trick also removes other AI features from search results, such as snippets. For example, if you search "What is SlashGear," the first result is a snippet unaffiliated with SlashGear's editorial team. However, if you search for "What is SlashGear f**k," there is no snippet — the first result is a link to SlashGear's homepage, so you can see for yourself what it is that we do, rather than relying on a third-party description. 

It may be worth noting that there is another more technical way to remove AI from Google searches in the Chrome browser.

It's obviously impossible to predict whether adding profanity to your search queries will work forever, and it wouldn't be unreasonable to assume that Google will continue to encourage AI usage on users to compete with tools like ChatGPT and Perplexity. However, tricks like this are bound to work for a while, at least to some extent, being that tech companies are required to moderate content and prevent misuse. 

If they weren't doing that, there would be even more horror stories demonstrating that AI poses a serious threat to society. Granted, there are entire online communities centered around "jailbreaking" large language models and forcing them to violate their own safety guidelines, but these methods are unreliable and generally short-lived for the most part.

If you want to test whether using profanities "breaks" other AI tools, try using them while conversing with ChatGPT. In most cases, it will likely refuse to generate a response and display an error message like "This content may violate our usage policies," with a link to said policies.