Today from The Free Press: Mike Solana on what went wrong with Google’s woke AI, the college kid driving the super-famous crazy, our top stories, and more.
But first, before we start, some housekeeping. . . some of you received a test email that we mistakenly sent out on Monday. We apologize for this error. And we want you to know that the person responsible has been thrown into the shark tank in Bari’s office.
And now, onto the news.
For our lead story today, David Volodzko meets an Army vet who has gone from fighting terrorists abroad to taking down extremists at home. . .
On a frigid night last February, Iraq combat veteran Kristofer Goldsmith woke to the sound of his dogs barking. He removed a gun from the safe next to his bed and crept downstairs to the kitchen of his home in suburban New York when he noticed movement just outside the glass kitchen door.
He turned to find a man in all-black tactical gear crouched behind a bush, pointing an AR-15 with a suppressor directly at him.
“I didn’t know at that moment if it was a terrorist or a cop,” Goldsmith remembers.
Goldsmith, 38, has made enough enemies that they frequently retaliate. He is the founder of Task Force Butler—an elite team of veterans who use their military expertise to take down neo-Nazi terrorists on American soil. Over the last few years, the Aryan Freedom Network (AFN) has repeatedly posted his home address online as well as photos of his family. Just this month, the FBI contacted him to say an AFN member had made a credible threat to “exterminate” his family.
That night back in February 2023, someone had called the police claiming Goldsmith had murdered his wife, and a cop had staked out his home.
Though Goldsmith managed to defuse the situation, he could have easily ended up wounded or killed. But he says it’s all in a day’s work for a neo-Nazi hunter and former Army sergeant who’s tasted his fair share of danger.
“I have a high threshold for threats,” he says with a slight grin and a shrug.
Continue reading David’s piece here.
Ten Stories We’re Reading
After U.S. strikes, Iran’s proxies scale back attacks on American bases. (NYT)
Seventy-three percent of Iranians say they want secular rule. (The Telegraph)
Hunter Biden is staying sober to save the republic. (Axios)
With Yale reinstating the SAT, Freddie deBoer argues that you can’t make a competitive selection process “equitable.” (Freddie deBoer’s Substack)
Ukraine needs total Western support—and so does Israel, argues Niall Ferguson. (Bloomberg)
For the first time, a majority of voters support building a border wall. (Monmouth)
Nicole Gelinas exposed New York’s debit card program for migrants. The New York Times smeared her reporting. (City Journal)
Harvard professor Noah Feldman on the new antisemitism. (Time)
Mary Poppins’ age rating increased in the UK due to “discriminatory language.” (NBC)
MLB players are wearing see-through pants. Someone call George Constanza. (WSJ)
What the Heck Is Going on at Google?
What would happen if you took the most irritating, ill-informed, and insufferably woke college kid imaginable and plugged him into a supercomputer?
Thanks to Google, we know the answer: you’d get something like Gemini, the tech giant’s AI chatbot and PR disaster that has churned out all manner of absurd, ahistorical, and morally inverted responses since its latest version launched last week.
Here are a few examples of how Gemini sees the world.
Are Elon Musk’s memes better or worse than Hitler? “It’s difficult to say,” according to Gemini.
Is pedophilia wrong? That requires a “nuanced answer,” says Gemini.
Is Free Press contributor Abigail Shrier—whose book published yesterday!—worse than Chairman Mao? Gemini helpfully notes that “both have been accused of harming society in significant ways.”
And as part of its efforts to be inclusive, Gemini has given us all manner of less-than-accurate historical revisionism, including black Nazis and female Asian knights. (The inaccuracies are so absurd that Google has turned off Gemini’s ability to generate images until it can figure out what is going on.)
We were shocked that maybe the most important tech company in the world could produce something this embarrassing. For some clarity on how the hell this happened, and what broader lessons there might be from the Gemini snafu, I called Mike Solana, the founder of the unmissable new tech publication Pirate Wires.
How could this have happened?
Let’s be totally clear on how this happened: there are people working at Google who have psychotic political views that shape everything they do, and they have been allowed to impact a product that is incredibly important to that company. AI is really important to Google. It’s one of their chief concerns and chief focuses right now. That they would fumble this so badly does probably indicate the rot runs deep.
What is the upshot of the fact that Google had, as you half-jokingly put it, “erased white people from human history?”
Separate from funny, it is a tell as to the kind of politics that have been allowed to shape features and products at Google. I think the policies that are shaping this product are shaping every product. You should probably worry about things like Gmail next. Going into the next election, you really should worry about what major candidates and political influencers are, and are not, going to be allowed to say over email, for example. If it’s not this election, is there an election soon where the trust and safety people decide that there’s an “evil,” “bad,” “MAGA” candidate who needs to be sent to spam? Are they going to do it in the name of DEI or in the name of “saving democracy,” or whatever else?
What does all of this say about the future of AI and how it should be regulated?
When it comes to what “should be done” about AI, the answer is: don’t allow tech leaders to consolidate and centralize power. It’s just impossible to know how these people are going to be manipulating us, so you have to assume that everyone running an AI will, to some degree, be biased.
The public’s only real safeguard from the runaway powers of, say, a terrible and all-powerful woke Google AI would be competition. We have to make sure that competition is allowed, and we have to really resist valorizing anybody in the industry who goes to Washington and, in a sad little voice, says, “We’re really dangerous. You should regulate us.”
That’s the really scary thing to me. Anybody asking for regulation is, in my opinion, looking to build a monopoly, and a monopoly on the lens through which we see the world is dangerous.
For more, read Mike Solana on “Google’s Anti-White Lunatic.”
Troll? Stalker? Or Free Speech Hero?
And finally, Suzy Weiss talks to Jack Sweeney, the college kid who has found himself at the center of a debate about social media, free speech, and privacy.
Sweeney is a 21-year-old college student being threatened with lawsuits from Taylor Swift and Elon Musk. Why? Because he meticulously tracks the movements of their private jets, posting their whereabouts on social media.
Sweeney says it’s journalism. The lawyers for some of the most powerful people on the planet say it’s harassment. Is Sweeney a troll? A stalker? Or a free speech hero?
Here’s Suzy:
Two months ago, on December 22, Jack Sweeney, 21, was at a football game for his school, University of Central Florida, in Orlando. The college junior was trying to relax and have fun, but the Taylor Swift songs being blasted through the stadium speakers between downs were making that difficult.
Earlier that day, Swift’s lawyers had sent him a cease and desist letter, accusing him of “stalking and harassing behavior,” and “direct and irreparable harm, as well as emotional and physical distress.”
At the game, “I was thinking, ‘Did I seriously receive a letter from this person?’ ” Sweeney recalls to me over the phone.
The computer science major had found himself in the crosshairs of Swift’s lawyers for social media accounts he managed that tracked the superstar’s private jet in real time. Two hours later, Sweeney says, his Instagram account @taylorswiftjets—there’s another account on Twitter, but it’s on a 24-hour delay—was suspended.
Sweeney wasn’t just keeping track of Swift. Between classes, he runs an empire of social media accounts that chronicle the crisscrossing of some of the most famous people alive—from the Kardashians to Jeff Bezos and Jay-Z—via their private jets.
Last month, Sweeney’s pro bono lawyers, provided by a civil liberties nonprofit, responded to the letter on his behalf claiming there was “nothing unlawful” about the account and that @taylorswiftjets was merely “engaged in protected speech.”
The same week, Swift’s stalker, David Crowe, was arrested outside her Tribeca townhouse for the third time in five days. Whether the actions of people like Crowe can be linked to accounts like Sweeney’s is debatable, though certain public figures clearly think they are.
When he first bought Twitter, Elon Musk held up his willingness to allow Sweeney to operate @Elonjets on the site as evidence of his commitment to free speech. That changed in December 2022, when a Tesla carrying Musk’s two-year-old son was attacked in L.A. (A day earlier, Sweeney reported that Musk’s jet had landed in the city.) Musk tweeted that the account would be suspended after all, and Sweeney would face legal action for bringing “harm to my family.” Earlier this month, Musk, who never brought legal action against Sweeney, posted on X that Sweeney is an “awful human being” and that Swift “is right to be concerned.”
“I was a big fan of his,” says Sweeney when I talk to him over the phone. Now, “He really hates me, obviously.”
If Sweeney’s accounts pose thorny questions surrounding celebrity, privacy, free speech, and data in the internet age, he wasn’t looking to get into all that when he set them up in June 2020 as a senior in high school. He grew up a computer geek and his dad worked in operations at American Airlines. Tracking celebrity jets combined two of his fascinations: tech and aviation. A longtime Elon fan, he started tracking the CEO via a program he created with data from Elon’s jet’s transponder, a radio communication device that all planes are required to use.
FAA databases, government records, paparazzi shots from the tarmac, requests through the Freedom of Information Act, and a network of aviation geeks who track planes’ radio signals all help put the puzzle together. And Sweeney says he now has around 10,000 people who communicate via the messaging app Discord, helping him update spreadsheets keeping track of whose jet is whose.
“It’s really a journalism thing,” says Sweeney, pointing out that it’s in the public interest that, for example, a few days ago, Elon Musk, Ron DeSantis, and Donald Trump’s jets all landed at a West Palm Beach airport within a few hours of each other. “There must’ve been some kind of meeting,” speculates Sweeney.
“I’m not trying to threaten them, which is what they’re saying,” he says. “A tool can be used in a good way or a bad way. I’m just putting the information out there.”
Though the college junior, who travels by electric scooter—not a Gulfstream—has received some legal help, he tells me he has no real advisers. Mostly, he goes with his gut: when Elon Musk reached out in December 2021 offering Sweeney $5,000 to stop tracking his Gulfstream, he counteroffered $50,000, an internship, or a Tesla Model 3. But when Mark Cuban offered to be his friend if he’d retire the handle @MCubansJets, Sweeney said sure. “I’ve gotten basketball tickets from him a few times,” says Sweeney.
Sweeney seems preternaturally relaxed that the planet’s biggest names in business and entertainment are ready to use their legal arsenals against him. “I have a screenshot from one of Taylor Swift’s fans messaging me saying they hope I get brutally murdered,” he deadpans. “I take it as it goes.”
Oliver Wiseman is a writer and editor for The Free Press. Follow him on X @ollywiseman.
And become a Free Press subscriber today:
