Unless you’ve been living under a rock, you likely have heard about the Epstein Files. The files show the various alleged crimes of Jeffrey Epstein, including sex trafficking of minors, and the people he allegedly corresponded with, including Donald Trump, Elon Musk, and former prince Andrew, who lost his title due to his supposed association.
Lawmakers passed a congressional act requiring Trump’s Department of Justice release the Epstein Files so lawmakers and the public could know who was associated with the notorious sex trafficker and potentially force new investigations into the allegations people made against alleged abusers in the files. However, when the DOJ released the files, they did so in small parts, and still may have not released all of them. Numerous files related to Donald Trump seem to have been withheld. They seemingly censored “Don T” and all instances of “don’t” as a result of trying to keep his name hidden. He appears in the files over 1,000 times anyway. The released version of the “Epstein Files” censored the name of alleged abusers, and yet somehow forgot to censor the names of the victims.
It seems Trump sycophants in the DOJ had clear goals.
Why abuse victims would want to sue the Department of Justice is obvious, but why Google? It seems Google, and specifically Google’s AI, made life difficult to the victims. A lawsuit claims Google made their names and information searchable. These victims who just wanted justice ended up becoming targets, and even after the DOJ finally redacted their names and information, far too late to prevent the damage, Google allegedly continued to make that information available. After multiple failed attempts to get Google to do the right thing, they sued.
DOJ Accused of Possible Retaliation
“In this context, DOJ’s decision to proceed with a massive, improperly redacted release constituted, at minimum, a willful and deliberate indifference to survivors’ privacy and, at worst, a punitive response to survivors’ public advocacy for transparency, effectively “doxxing” victims who spoke out by exposing their identities and contact information.”
– from the lawsuit against Google and the U.S. Government
The Epstein files mention Donald Trump more than 1,000 times. Trump had a well-documented friendship with Jeffrey Epstein, and the files have proven contentious for Trump and his allies. The people who pushed for their release in congress, such as Marjorie Taylor Greene, once an ally of Trump’s, found themselves on Trump’s bad side.
It’s rational to worry that Trump and his ardent supporters in the DOJ would be more focused on protecting Trump than the victims who also pushed for the release of the documents. With the DOJ flubbing up the incredibly simple task of censoring files and removing the names of victims, it feels almost ridiculous to claim that so many people involved with the release of these documents all made the same mistakes and missed this data. It’s easier to assume it wasn’t a priority for the DOJ to protect the victims, not the accused perpetrators. I usually say don’t ascribe to malice what can be explained with incompetence, but that would imply a lot of people in Trump’s orbit are incompetent.
Google Makes Personal Data Accessible
The DOJ opened a can of worms, but Google may have spread them everywhere. While the DOJ endangered the alleged victims of Epstein and his associates by releasing their names and contact information, Google, according to this lawsuit, republished the personal information, made it searchable, and even generated a hypertext link that allowed someone to send an email to an Epstein victim with a click.
“Survivors now face renewed trauma. Strangers call them, email them, threaten their physical safety, and accuse them of conspiring with Epstein when they are, in reality, Epstein’s victims.”
– from the lawsuit against Google and the U.S. Government
The information Google allegedly made widely available through their AI Mode includes the victim’s “full name, contact information, cities of residence, and association with Jeffrey Epstein.” According to the victim at the center of the suit, they tried to get the information taken down, and Google did nothing. This also makes Google’s AI unique, as, according to the lawyer representing the victims, ChatGPT, Claude, and Perplexity all did not give the victim’s information to users requesting it.
“Despite receiving actual notice of the violations, the substantial harm caused by its continued dissemination, and the status of many Class members as sexual abuse survivors entitled to heightened privacy protections under the law, Google has failed and refuses to remove, de-index, or block access to the offending materials.”
– from the lawsuit against Google and the U.S. Government
It’s worth noting that this isn’t a part of Google’s search feature. They didn’t index the Epstein documents and make them more searchable. Instead, the ingested the data from the documents and provided it through AI. The plaintiffs hope this means it will count as “actionable doxxing,” a stance potentially strengthened by the claim that Google had not changed the functionality to protect vulnerable victims after they requested that Google remove their personal information.
This is what happens when you don’t control your datasets or put human safety first. It’s a problem at the core of nearly all AI. If you’re not controlling the data you’re adding to your dataset, you can never fully control the output. You will add information you don’t want the AI spitting out. That can include copyrighted material, original artworks, and the complete personal details of victims of a child sex trafficking ring. These are people who were victims of powerful people. It’s a list that includes the President of the United States and world’s richest man, who have plenty of antagonistic fans who would jump at the chance to harass Epstein’s victims.
No wonder they want to let AI operate without restrictions. Creeps of a feather flock to AI.