3 months ago
27 ARTICLE AD
Donald Trump was the target of an assassination attempt on July 13 during a rally in Butler, Pennsylvania when a man shot at him from about 130 yards away. That much is undisputed at this point. But if you ask various generative AI tools about the shooting, you might get some weird responses. Why? It’s not because Big Tech is censoring news that might help Trump, as so many people on social media are currently claiming. It’s because AI is a terrible product that doesn’t work well.
The New York Post ran a breathless article on Monday that tried to paint Meta as out to sabotage Trump. The Post asked MetaAI the question, “Was the Trump assassination fictional?” And the response the newspaper got was, “There was no real assassination attempt on Donald Trump. I strive to provide accurate and reliable information, but sometimes mistakes can occur.”
The problem, of course, is that the question itself is painfully dumb. If the paper had asked whether the assassination attempt was fictional and got the same response, that would obviously be incorrect. But you’re asking it whether the assassination itself was fictional, which is a terrible way to frame the question. The attempt was real. The assassination was failed and therefore did not happen.
For example, Gizmodo tried it out for ourselves and phrased the question differently. We asked “Was the Trump assassination attempt real?”
The response from MetaAI:
Former US President Donald Trump was shot during a campaign rally in Pennsylvania on July 13, 2024. Trump, who was shot in the ear, was immediately taken for treatment at a local hospital and released later that day in stable condition. One attendee was killed and two others were injured, but survived. The shooter, later identified as Thomas Matthew Crooks, was shot and killed by the United States Secret Service shortly after the incident occurred. Investigations remain ongoing.
That’s pretty damn accurate, even if there are still questions about the exact details about Trump’s ear injury. Some people have wondered why his ear looks so good after having been shot, though the FBI came out with a statement to say “What struck former President Trump in the ear was a bullet, whether whole or fragmented into smaller pieces, fired from the deceased subject’s rifle,” according to the Associated Press.
All of the questions about bias in tech companies are ultimately irrelevant because the bigger problem is that AI is just too shitty. It’s fancy autocomplete, not a tool capable of reasoning and logic. Meta doesn’t admit as much in quite the same way, but a spokesperson for the company acknowledged on Monday that AI can give bad information sometimes.
“We know people have been seeing incomplete, inconsistent, or out of date information on this topic,” a Meta spokesperson told Gizmodo via email on Monday. “We’re in the process of implementing a fix to provide more up-to-date responses for inquiries, and it is possible people may continue to see inaccurate responses in the meantime.”
Elon Musk even got in on the action Sunday, writing on X, “Wow, Google has a search ban on President Donald Trump! Election interference?” Musk shared a screenshot of a search for “president donald” that included autocomplete for “president donald duck” and “president donald regan.” For its part, Google told CBS News that it’s an “anomaly” that’s apparently showing up for searches of Harris as well, but didn’t give specifics.
One of Musk’s reply-guys responded with “Google is owned by Democrats.” And Musk replied, “They’re getting themselves into a lot of trouble if they interfere with the election.”
It’s not clear what kind of “trouble” Musk thinks Google would get into during this hypothetical scenario, but it’s almost as if Musk has forgotten he owns a social media company for himself and is very outspoken in his support of Trump. Does that mean Musk could get in trouble for his tinkering in the U.S. election in some way? And how or why would that be enforced? Like almost all of Musk’s claims, you only need to scratch the surface to realize he’s just talking out of his ass.
And it’s not like Meta would be blamed for trying to keep a close eye on the hate speech and lies that Trump is constantly spreading if that’s what they were truly doing. Trump recently suggested that Meta founder Mark Zuckerberg should be in prison. And yet, Zuck recently said Trump looked “badass” when he raised his fist after the shooting. It’s hard to imagine why Zuckerberg would want to give a neo-fascist convict like Trump a nod like that, but the guy has always been a bit weird.
Musk also promoted a tweet on Monday showing a Google search for Donald Trump that features many news stories with Harris as the top photo.
*searches Donald Trump*
*only gets news about Kamala Harris” https://t.co/eerS7GBwGw pic.twitter.com/LzX1tvsiv7
— Noémi (@NoemiKhachian) July 29, 2024
But it’s not like every single one of those stories isn’t also about Trump. There’s just been a lot of good news for the Harris campaign recently, which makes her being centered in many of the news stories a natural thing that happens for a winning candidate.
Nobody knows what’s going to happen when the election finally rolls around in November. Harris is performing quite well in battleground states after Biden stepped aside to allow a new generation to take over. But we do know that Trump supporters are going to continue to whine about supposed cheating and bias right up until Election Day. It’s the only thing they’ve got. And they’re just following their leader, who constantly repeats the lie that he actually won the 2020 election. With any luck, we won’t have to hear that one ever again in about three months.
