Google nonetheless recommends utilizing pizza glue

I agree, the “how a lot glue to placed on pizza” query is uncommon, however not that uncommon given the current fuss about glue pizza. As famous by Colin McMillen on Bluesky, in case you ask Google how a lot glue to placed on a pizza, the right reply is not any! – doesn’t seem. As a substitute, it quotes our woman Katie suggesting including an eighth of a cup. Oops!

You may be questioning if this can be a pretend screenshot. I puzzled this too. However Edge confirmed by working our personal question:

Guys, that is simply phenomenal stuff. Each time somebody like me reviews that Google’s AI is doing one thing fallacious, we practice the AI offender.

These of us, ahem, sure age will bear in mind the phenomenon “Google bombs;” a traditional instance was using the phrases “depressing failure” in reference to George W. Bush. This was executed usually sufficient {that a} Google seek for “depressing failure” would return George W. Bush. In some unspecified time in the future within the late 2000s, Google discovered easy methods to crush this enjoyable recreation, however due to new AI outcomes, hey, the sport is working once more! I am simply going to put in writing “pathetic failure” in the identical sentence as George W. Bush once more for outdated occasions sake, and perhaps in a day or two you will get a terrific new AI search outcome, who is aware of!

This, by the best way, will not be a common drawback. I requested Perplexity.AI how a lot glue to placed on a pizza and he instructed me, “I might strongly advise in opposition to placing glue on a pizza. Glue will not be an edible ingredient and consuming it could be poisonous and dangerous to your well being.” The next explains how the “glue on pizza” meme happened.

ChatGPT additionally doesn’t advocate sticking glue on pizza:

Naturally, this isn’t the one factor that goes fallacious, though it’s maybe probably the most amusing. However this is one thing else that is fairly good: Google can now not reply questions on its merchandise due to its synthetic intelligence. Verge editor Richard Lawler requested easy methods to allow screenshots in Chrome incognito mode. Google’s AI gave two solutions, each fallacious. One in every of them suggests taking a screenshot of a daily Chrome tab.

In one other case, Google’s AI insists that taking a screenshot in Chrome’s incognito mode is solely unimaginable:

Sadly, by describing this drawback, I’m fairly positive that I’m making it worse. Google goes to swallow my glorious description of the issue and return it to the unwary as proof that Chrome screenshots in incognito mode are unimaginable and that the glue belongs to your pizza. I ponder what mischievous bloggers will do with this data?