Grok AI Creates Bizarre Fake News About the Solar Eclipse Thanks to Jokes on X – Gizmodo

author
3 minutes, 58 seconds Read
Screenshot: Gizmodo

Grok, the AI tool available to paying customers on X, generated another fake news story based on jokes Monday. And while this latest story isn’t as dire as last week’s inaccurate “news” that Iran had launched an attack on Israel, it still proves Grok has a long way to go before it’ll ever be a useful and accurate tool.

The headline available to Grok subscribers on Monday read, “Sun’s Odd Behavior: Experts Baffled.” And it went on to explain that the sun had been, “behaving unusually, sparking widespread concern and confusion among the general public.”

Advertisement

The news item claimed that even with the public’s growing concern, “there has been no official statement or explanation from scientific authorities regarding the sun’s current state.” Needless to say, scientists definitely understand what was happening with the solar eclipse, which was observed by millions in North America on Monday. There’s no mystery about why an eclipse happens, despite what some fringe U.S. politicians may insist.

Advertisement

The tweets that apparently prompted this bizarre proclamation from Grok included joke after joke of people wondering where the sun had gone. It appears that so many people making the same joke confused Grok, which tried to turn the jokes into something we should all be worried about.

Advertisement

“WHAT THE FUCK IS HAPPENING TO THE SUN RIGHT NOW?????” the top tweet, written by comedian Michael Ian Black, read on Monday.

“Something’s going on with the sun. Don’t want to alarm anyone but seems like I’m not the only one noticing it,” another X user joked.

Advertisement

Grok, which was first launched to a limited batch of users in December but is now available to all paying customers, has been touted by X owner Elon Musk as the “anti-woke AI.” Musk has repeatedly complained that competing artificial intelligence tools are too “woke,” a sentiment he shared yet again during an X Spaces chat on Monday. In fact, Musk claimed that woke AI may just kill off real humans in the future because it’s trying to enforce some diversity quota.

“You can take the Google Gemini example where it refused to publish… to produce a picture of George Washington as a white man,” Musk said on Monday.

Advertisement

“And any historical figure would automatically be made diverse because it’s been programmed to insist on diversity, which sounds, you know, perhaps okay at first. But not if the AI has so much power that it can actually enforce diversity and decide there’s too many of one kind of people or too many of one sex and just kill off enough until the diversity number is what is programmed to believe is correct,” Musk continued.

Whatever you think of Musk’s take on the politics of AI, Grok has stumbled again and again by creating fake news stories out of tweets from X users, including last week when it generated the headline, “Iran Strikes Tel Aviv with Heavy Missiles.” Israel has been on high alert in anticipation of a potential attack by Iran, with Israel even scrambling GPS signals in a bid to confuse any incoming missiles. But there has been no attack on Tel Aviv, despite what Grok insisted.

Advertisement

Rest assured, Grok isn’t making these mistakes because it’s too woke or too anti-woke. AI commonly makes mistakes because it’s simply not very good technology yet. AI boosters have spent the better part of the last year convincing the general public that it’s a miracle advancement that will soon achieve something close to human sentience. But don’t hold your breath on that last part.

Generative AI is little more than a souped-up version of autocomplete, which means it can sound pretty convincing as it mimics the ways in which humans talk. But it has no ability to apply complex reasoning or logic. It’s simply remixing words in an admittedly fast and impressive fashion. But the “brain” behind this tech isn’t thinking, even when it appears to be.

Advertisement

Grok didn’t think about the tweets it saw of people joking about the solar eclipse. It identified patterns on X and spit out human-sounding text summarizing those patterns. And this is just going to keep happening because the only way to correct it is with enormous armies of humans working as mechanical turks behind the scenes. Grok isn’t dumb because it’s woke or anti-woke. It’s dumb because it’s pulling off a magic trick that countless people are falling for.

This post was originally published on this site

Similar Posts