Fucksmiths, Google, and the Rise and Fall of Artificial Intelligence in Search & Measuring: How Google I/O Made the Pizza Glue

Imagine this: you’ve carved out an evening to unwind and decide to make a homemade pizza. You are ready to eat your pie after putting it in the oven. But once you get ready to take a bite of your oily creation, you run into a problem — the cheese falls right off. You are frustrated, so you turn to the Internet for a solution.

Don’t do that. As of writing this, though, that’s what Google’s new AI Overviews feature will tell you to do. The feature isn’t triggered for every query, but it does check the web and produce anai-generated response. The answer received for the pizza glue query appears to be based on a comment from a user named “fucksmith” in a more than decade-old Reddit thread, and they’re clearly joking.

It’s an odd situation, since Google has been testing AI Overviews for a year now — the feature launched in beta in May 2023 as the Search Generative Experience — and CEO Sundar Pichai has said the company served over a billion queries in that time.

Take Google I/O’s big launch of this feature, for instance. The demo was controlled and it delivered a questionable answer about how to fix a jammed film camera. (It suggested they “open the back door and gently remove the film”; don’t do that unless you want to ruin your photos!)

Openai, Meta, and Perplexity all have grappled with the effects of artificial intelligence. There are examples of flubs all over the place, and that’s because it was the first to use this technology on such a large scale.

Companies developing artificial intelligence are often quick to avoid taking accountability for their systems with an approach much like a parent with an unruly child — boys will be boys! These companies claim that they can’t predict what this AI will spit out, so really, it’s out of their control.

Our search experiences are being affected by posts from 10 years ago about incorporating artificial intelligence into everything. Many idealists say that we are on the verge of something amazing and that the issues are simply the growing pains of a new technology. I sure hope they’re right. Something is certain, we will probably witness someone putting glue on a pizza soon.

The artificial intelligence optimist argues that because of the rapid progress made so far, we should embrace the hype. I believe that this technology will continue to get better, but focusing on an idealized future where these technologies are flawless ignores the issues they currently face, and allows companies to continue delivering subpar products.

Are AI Models Really That? An Analysis of AGI’s Misleading AI Contribution to Google Search and Memocracy Disable AI Overviews

Marcus said that the models are constitutionally incapable of sanity checking on their own work, which has bitten the industry in the behind.

Today’s feature is just a small part of what the company announced last week. There is a lot of ambition here for the ability to create an artificial intelligence organized results page, as well as for multistep reasoning for complex queries. Right now the company’s reputation hinges on just getting the basics right and it isn’t looking great.

“You actually need to do some reasoning to decide: is this thing plausible? Is this source legitimate? You have to do things like a human fact checker might do, that actually might require artificial general intelligence,” Marcus said. And Marcus and Meta’s AI chief Yann LeCun both agree that the large language models that power current AI systems like Google’s Gemini and OpenAI’s GPT-4 will not be what creates AGI.

“A company once known for being at the cutting edge and shipping high-quality stuff is now known for low-quality output that’s getting meme’d,” one AI founder, who wished to remain anonymous, told The Verge.

But Pichai has also said that Google’s brought the cost of delivering AI answers down by 80 percent over that same time, “driven by hardware, engineering and technical breakthroughs.” It appears that kind of optimization might have happened too early, before the tech was ready.

Some examples of the weird things that a new product from the search giant will say are being shared on social media. The messy rollout means Google is racing to manually disable AI Overviews for specific searches as various memes get posted, which is why users are seeing so many of them disappear shortly after being posted to social networks.