Timnit Gebru had expected her colleagues to rally around her when she was abruptly fired from Google on December 2nd. She was a well-respected AI ethics researcher, her termination as controversial as it was sudden. What she hadn’t anticipated was becoming a catalyst for labor activism in Silicon Valley — or the subject of a harassment campaign that surfaced alongside her supporters.
Her firing came weeks after Google managers asked her to retract a paper on the dangers of large language models, like the ones that power the company’s search engine. Gebru pushed back, saying the company needed to be more transparent about the publication process. Employees saw the termination as a blatant act of retaliation, and thousands of workers, researchers, and academics signed an open letter demanding an explanation.
In early December, as media attention mounted and tech workers across the country came to her defense, Gebru waited to see how Google would respond. What is this company capable of? she asked herself. She didn’t have to wait long to find out.
Twitter enabled Gebru’s colleagues to stand up for her; it also made her vulnerable to a small but very active group of harassers
On December 4th, Jeff Dean, Google’s head of AI, published his views on Gebru’s dismissal. He said she had resigned — a fact she openly disputed — and that her paper hadn’t met the bar for publication. His words lay the groundwork for a very different type of campaign, one which used the groundswell of support surrounding Gebru’s firing as a stand-in for the ills of cancel culture. Twitter enabled Gebru’s colleagues to stand up for her; it also made her vulnerable to a small but very active group of harassers.
Over the next two months, Pedro Domingos, a professor emeritus at the University of Washington, and Michael Lissack, a Wall Street whistleblower turned admitted harasser, promoted the narrative that Gebru’s work was “advocacy disguised as science.” They said she’d created a toxic environment at Google and was “obsessed with being a victim.” They wrote off her supporters as sycophants and “deranged activists.”
On Twitter, the campaign gained momentum with the help of anonymous accounts. Gebru suspected they were sock puppets, fake profiles created for the sole purpose of harassment, since many had popped up around her firing. Some claimed to work in tech, using phrases like “pro-DEI minority,” “She/them,” and “LBGTQ Latinx” in their bios. When Emily M. Bender, a computational linguistics professor at the University of Washington and Gebru’s co-author on the paper, reported one of these accounts for harassment, Domingos implied she was being prejudiced.
“He enables these people, and gets to distance himself because he’s not the one saying ‘go back to Africa’”
To Gebru, the harassment wouldn’t be possible without Jeff Dean, who stayed quiet about the campaign for months, despite being tagged in numerous posts. “He enables these people, and gets to distance himself because he’s not the one saying ‘go back to Africa,’” she says. “Pedro [Domingos] is smart enough not to say these types of things, too. They hide behind civility and enable the trolls. That’s how they get away with it.”
Neither Domingos nor Lissack had an obvious connection to Gebru’s work prior to December 2020. But Domingos had a history of battling those he saw as “militant liberals” — particularly in academia, where he believed free speech needed to be protected. Lissack’s connection, on the other hand, was more random. He became fixated on proving that Gebru’s work wasn’t science, though he himself is not an AI expert. The men met over email and collaborated through social media. Where they intersected was the belief that Gebru was trying to cancel them and that they were, improbably, the real victims.
When Domingos first saw the news about Gebru’s firing, he didn’t think to intervene. He was dealing with another battle: the world’s largest machine learning conference was requiring presenters to submit a statement on the impact — good and bad — that their research could have on society, according to Nature. “It’s alarming that NeurIPS papers are being rejected based on ‘ethics reviews,’” he tweeted. “How do we guard against ideological biases in such reviews? Since when are scientific conferences in the business of policing the perceived ethics of technical papers.”
It was part of an ongoing fight in the academic community. In 2018, conference attendees, led in part by machine learning expert Anima Anandkumar, pushed to change the acronym of the event, “NIPS,” because it was “prone to unwelcome puns.” Anandkumar cited an unofficial pre-conference gathering called TITS and T-shirts that read “my NIPS are NP-hard.” The debate resurfaced in 2020 when Domingos said “the name change was a terrible idea” that Anandkumar had bullied the conference board into accepting.
Domingos had been interested in similar issues at the University of Washington. In June 2018, Stuart Reges, a faculty member at the school, wrote an article in Quillette, the “voice of the intellectual dark web” titled “Why Women Don’t Code.” In it, he expressed support for James Damore, a Google engineer who was fired after circulating a memo that argued biological differences between men and women might explain the gender gap in tech.
“Perhaps there aren’t more women in CS because they’re not interested, and we shouldn’t force them to be”
Reges’ article upset some students, and the computer science department sent out an email saying it disagreed with his claims. Domingos responded to the graduate student mailing list, writing that, “One of Stuart’s (and Damore’s) main points is that perhaps there aren’t more women in CS because they’re not interested, and we shouldn’t force them to be. We can press on regardless, but whether that’s best for either women or men is a question that surely deserves consideration.” He added: “When a woman is hired or admitted over a more qualified man, the man was discriminated against. Referring to general notions of oppression, benefits, etc. doesn’t make it legal or ethical.” One student responded that this type of comment was helping to create a “hostile and unwelcoming environment” in computer science.
In December 2020, when Domingos tweeted about the NeurIPS ethics review, Anima Anandkumar jumped in, saying the community needed to get rid of extremists like him in order to be able to thrive. She then published a “block list,” which included people who’d liked his posts to “establish accountability on social media.” The list prompted a backlash from people who felt they’d been unfairly included. Domingos said he was reaching out to her employer to try to curtail her “disgraceful behavior.” Anandkumar deleted the list, published an apology on her blog, and shut down her Twitter account. (Anandkumar did not respond to a request for comment from The Verge.)
To Domingos, the scuffle indicated why it was important to fight even small battles against what he classified as the “woke mob.” He’d grown up in Portugal, a country that transitioned from a dictatorship to democracy in the 1970s. “Each one of these things sounds very innocent, but then it sets up the stage for the next step,” he tells The Verge. “Once I get to declare what speech is harmful and what isn’t, I can install my tyranny because I can decide what you say and not.”
In the midst of the spat with Anandkumar, Domingos began tweeting about Gebru, echoing Jeff Dean’s claim that the researcher hadn’t been fired. “Timnit’s paper was not censored...and she was not fired,” he wrote on December 11th. “Rather, if you trash your company’s efforts in your area and threaten to resign, you shouldn’t be surprised if they let you.”
Lissack later pleaded guilty to harassing executives at the firm
His view was shared by an unlikely ally: a whistleblower named Michael Lissack who’d worked at Smith Barney in the ‘90s and exposed “widespread fraud in the municipal bond market,” according to The New York Times. The move made Lissack somewhat of a folk hero in the financial services industry. Lissack later pleaded guilty to harassing executives at the firm, sending prank emails and soliciting phone calls that tied up the institution’s main phone line.
After Smith Barney, Lissack continued to fight against issues he found “morally offensive” — though his definition of this seemed to evolve. In 2017, he filed an official grievance against Phaedra Parks, an entertainment lawyer and Real Housewives of Atlanta star, to try to get her disbarred. He also claims to have exposed Vicki Gunvalson, of the Real Housewives of Orange County, for running a fake cancer charity.
On his resume, Lissack said he’d been a visiting research professor at George Washington University. When The Verge reached out to the school to confirm, a media relations specialist said he’d checked with multiple groups but could not verify Lissack’s employment. Lissack forwarded an email from a professor at the school, who said he’d been a “guest lecturer.”
In December 2020, Lissack read a copy of Gebru’s paper that had leaked on Reddit. He sent Domingos an email with his assessment, though the two did not know each other personally:
I read the draft ... not academically sound
Lack of declaration of standards of evaluation.
Lack of declaration that this is really an abductive OPINION piece.
Lack of adequate discussion of trade-offs.
No disclosure of competing hypotheses.
Then he decided to publish an article laying out his views. He said Gebru’s paper was, at best, sloppy science, and failed to acknowledge that it was a “position paper / advocacy piece rather than research.” He even created a Twitter account to promote his work.
Domingos quickly stepped in to help. “The paper by @timnitGebru et al that led to her resignation is advocacy disguised as research, says philosopher of science @LissackMichael in this devastating critique,” he wrote.
Over the next two weeks, a familiar pattern emerged. Gebru would tweet something related to her firing, or Google, or AI ethics research, and Lissack would jump in with a response, often tagging Domingos. On January 27th, Gebru wrote that some of her critics were making up arguments that weren’t in her original paper. “How would you know since you will in general only politely interact with the sycophants,” Lissack responded. “Everyone else you call names and mistreat and then falsely claim you are a victim…@pmddomingos.”
The next day, Gebru tweeted about April Christina Curley, a diversity recruiter who was fired from Google in September, and alluded to her own termination. “You were ‘resignated’ for insubordination attempted sedition and attempted extortion,” Lissack wrote. “You DID NOT do your job.” He also said that Gebru “does not belong working anywhere as a corporate employee.”
This opinion was echoed by anonymous Twitter accounts that claimed to belong to Google employees. One, which used the name Jeff Jeffries, said they worked at Google Brain and used a photo of a Black man in their avatar photo. Another, Bili Franklin, said they were a product manager at Google, used they / them pronouns, and identified as Latinx. (A Google spokesperson confirmed that neither of these people works at Google.)
On December 31st, “Bili Franklin” tweeted screenshots of an anonymous employee forum called Blind, where a Google worker had written that Gebru had “rancid egomaniacal outbursts” and was unstable. (Blind assumes someone is a Google employee if they have a Google email address.) Domingos retweeted the screenshots. “For anyone who was wondering, this is what people inside Google think of @timnitGebru’s antics,” he wrote.
Bender reported the tweet for harassment, prompting Domingos to respond: “Wow. Emily Bender, a supposedly respectable academic and defender of minorities, is calling @BiliFranklinBLM, a Latinx LGBTQ PM at Google, a sock puppet just because she dares to criticize @timnitGebru.”
More anonymous Twitter and email accounts popped up. Some called Gebru a bitch and told her to “go back to Africa.” Others said she was arrogant and only hired because she is Black.
“He wanted to catch me”
When Gebru pushed back at the men on Twitter, it only seemed to inflame them further. Once, she cracked and told Domingos to go fuck himself. “This is how an AI ethicist speaks to a fellow computer scientist,” he responded. Gebru felt like she’d fallen into his trap, though she didn’t regret what she’d said. “That’s what he wanted,” she adds. “He wanted to catch me.”
In fact, Domingos had laid out his tactics for combating “social-justice radicals” in an article in Quillette. “Goad [the cancel crowd] into overreaching,” he said. “Mock them mercilessly.” He added that “cancelers take themselves extremely seriously, imagining themselves to be social-justice angels whose holy ends justify every imaginable means. Their sanctimonious spirit is a gift to you, if you call it out instead of playing along with its conceit.”
In January, a PhD student at the University of Washington tweeted parts of the article on Twitter. Lissack then posted her diversity statement where she spoke about her struggles with mental illness. “[The student] acknowledges her mental illness here,” he wrote. “Much better to disengage Even better would be for her to get help.”
Lissack later told The Verge the tweet was “an error,” though he did not stop going after Gebru’s supporters. In February, he emailed the University of Washington to report “research misconduct” by Emily M. Bender and wrote an open letter to Margaret Mitchell, co-lead of Google’s ethical AI team, who was fired after using an automated script to look through her emails for evidence of discrimination against Gebru. He also wrote Alex Hanna, a research scientist on Gebru and Mitchell’s team, an email saying she was “next” and told her colleague, Emily Denton, that their tweets had “crossed a line.” When Mitchell tweeted that she’d lost her health insurance, he emailed her former co-workers to suggest she sign up for Obamacare.
Lissack also contacted a woman who’d been collecting screenshots of his activity on Twitter. “Karma is not a nice lady,” he wrote. “You have been warned. She is very patient and highly effective.” Lissack emailed the woman’s husband to say that his patience had ended.
When asked about the threatening notes, Lissack said the woman had “organized the campaign to have me banned on Twitter. It had to stop.”
On February 1st, Lissack was suspended from Twitter. He’d tweeted roughly 2,000 times in the 17 days he had his account, mostly about the controversy surrounding Gebru.
In an interview with The Verge, Lissack says he wasn’t sorry to leave the platform. “There is this horrible pull that people have on Twitter,” he explains. “Why bother getting into nuance? The net result is that people get sucked into directions they might not actually wish to go. I’m guilty. It happened. I’ll learn from it.”
“Karma is not a nice lady”
When asked why he’d gone after a Real Housewives star in 2017, Lissack responded, “You haven’t talked to many whistleblowers. I see something that to me is morally offensive. People may not agree with the things that strike me as wrong but I am who I am, what can I tell you. To me, the media representing [Gebru’s paper] as good science is offensive.”
On March 1st, Jeff Dean finally spoke out about Lissack’s behavior. “I (and Google) have no connection to Michael Lissack,” he wrote on Twitter. “I have seen enough disturbing reports of this sort of behavior that I have emailed him and asked him to cease contacting people who don’t want to be contacted. This kind of behavior has no place in scientific discourse.”
Lissack’s theory about Gebru has evolved since The Verge started reporting this story. He now compares her to Aaron Swartz, a programmer and activist who died by suicide in 2013. He says Bender, a renowned researcher, has been pushing Gebru to “not let go” for her own gain. “The ONLY person benefiting here is Emily,” Lissack wrote in a text message to The Verge. “Emily is becoming influential when before this she was a no body.”
On February 9th, Twitter unsuspended Lissack’s account. Domingos welcomed him back, writing, “Glad to have @LissackMichael back after the cancel crowd’s failed attempt to expel him from Twitter.” Lissack made his account private, and changed his bio to say “UNCANCELLED!!”
“I live in Salem, Massachusetts,” he tells The Verge. “What occurred on Twitter was no different than a version of the witch trials.” It’s perhaps telling that Lissack sees himself in the role of unjustly persecuted woman, rather than misogynistic male judge.
Lissack later deleted his account, saying the “17 day adventure” was “official over.”
To Gebru, the strategy was laughably transparent. “The funniest part to me is the cancel culture stuff,” she says. “Who is fired? It’s me. But now they’re complaining about cancel culture. I didn’t even know about Pedro Domingos or his work before he started tagging me on Twitter.”