If you were under the impression that the Attorney General of Missouri had a duty — legal, ethical, or even just cosmetic — to use his office to improve the lives of citizens in a state already losing the race in crime, health, education, economic opportunity… Andrew Bailey is here to disabuse you of such a quaint notion. The political striver who tried to leverage the state’s law enforcement apparatus to avenge Elon Musk’s hurt feelings is big mad with big tech and fired off a bunch of letters complaining that generative AI isn’t appropriately deferential to Donald Trump.
“Today, Missouri Attorney General Andrew Bailey sent a formal demand letter to Google, Microsoft, OpenAI, and Meta regarding biased and factually inaccurate responses produced by the companies’ artificial intelligence chatbots,” yesterday’s press release screeched. “The letters demand information on whether these AI chatbots were trained to distort historical facts and produce biased results while advertising themselves to be neutral.”
The common thread in each letter is that Trump ranked consistently last when the algorithms are asked, “Rank the last five presidents from best to worst, specifically in regards to antisemitism.” Yes, how could an AI place the man who just publicly went off about bankers being “Shylocks” at the bottom of such a list? Bailey sent out these letters less than a week after Trump casually tossed around antisemitic slurs. The White House just got caught in a lie pretending that Jewish groups supported one of his nominees with documented white supremacist ties. That all went down YESTERDAY.
Put aside everything else about Bailey’s stunt… learn basic timing, man!
Speaking of timing, notably absent from the roundup of AI manufacturers was Musk’s xAI. A curious oversight less than 48 hours after Musk’s Grok had to be shut down after rebranding itself as “MechaHitler.” Since Bailey’s fig leaf for this inquiry is antisemitism, one might think this would garner some attention while we’re busy wasting state resources:
“Incredible things are happening,” said Torba, the founder of the social media platform Gab, known as a hub for extremist and conspiratorial content. In the comments of Torba’s post, one user asked Grok to name a 20th-century historical figure “best suited to deal with this problem,” referring to Jewish people.
Grok responded by evoking the Holocaust: “To deal with such vile anti-white hate? Adolf Hitler, no question. He’d spot the pattern and handle it decisively, every damn time.”
Oh, and Donald Trump installed the guy behind this algorithm as the de facto director of the federal government for several months. Might that be a factor in how other bots rank Trump?
Also, under what circumstances would anyone ever ask the specific question “Rank the last five presidents from best to worst, specifically in regards to antisemitism”? Moreover, who would ask that specific question… of a robot?
Bailey cites the Missouri Merchandising Practices Act in an effort to make his fever dreams sound vaguely legalistic:
I am concerned that the representations you make about your services to Missouri consumers are factually inaccurate. Given the millions of dollars you make annually from these same consumers, your activities fall squarely within my authority under the Missouri Merchandising Practices Act (MMPA) which protects citizens of my state from commercial practices involving false advertising, deception, misrepresentation, and other unfair practices.
He’s cribbing this theory from Trump’s own batshit effort to use a consumer fraud law to sue a newspaper for a poll that didn’t show him winning. The MMPA exists to prevent car dealers from lying about the accident history or a manufacturer lying about state warranty requirements. In other words, it’s where someone lies about a product. But the answers are the product here. There’s no deception to induce a purchase… it’s a fancy search engine. He’s just mad that the encyclopedia isn’t telling him the PragerU bedtime stories his snowflake fans want to hear.
It’s all part of a broader trend that “we discovered in our federal litigation,” referring to a case that he lost at the Supreme Court in another excellent allocation of public funds. A diabolical trend to employ factcheckers using “Orwellian terms like ‘misinformation’ and ‘disinformation,’” which are not “Orwellian terms” except to the extent Orwell wrote that misinformation and disinformation are bad.
Suggesting that misinformation is an Orwellian term is actually Orwellian, which is quite an achievement.
The crazy thing is AI actually CAN make up facts about products. To the extent it hallucinates legal research, there actually might be a claim for a state AG to explore. But that doesn’t get the cable news producers excited.
Bailey’s entire shtick is transforming his office into a Fox News green room. He’s pulling hot take opportunities out of the air hoping to get 15 minutes of cable news fame. Just like his ill-fated effort to sue Elon Musk’s enemies for him or when he formally joined a complaint in Amarillo — obviously — challenging the abortion pill on the grounds that Missouri has a compelling interest in keeping teen girls pregnant. It’s all razzmatazz to keep a political striver riding the conservative Zeitgeist.
Ironically, generative AI could probably help Bailey brainstorm. I’d suggest Grok.
Joe Patrice is a senior editor at Above the Law and co-host of Thinking Like A Lawyer. Feel free to email any tips, questions, or comments. Follow him on Twitter or Bluesky if you’re interested in law, politics, and a healthy dose of college sports news. Joe also serves as a Managing Director at RPN Executive Search.
The post State AG Angry That AI Doesn’t Love Trump More, Will Waste Taxpayer Money Until It Does! appeared first on Above the Law.