Dylan Ventura is not a Luddite. As a part of the new UC Berkeley chapter of QuitGPT, Ventura isn’t an AI zealot either.
I met Ventura on the back patio of Caffè Strada during one of those hot Berkeley afternoons in early March. Ventura, a sophomore studying economics, is a part of the QuitGPT campaign’s newly-initiated — a group of undergrads bringing the national organization’s movement to UC Berkeley’s campus.
Ventura explained his motivations as we sat in the shade of Caffè Strada’s green awning: “QuitGPT is the campaign and affiliated organization focused on getting people to stop using ChatGPT because of the moral and ethical implications associated with their CEO and president,” Ventura told me. He referenced OpenAI’s recent backdoor deal with the Pentagon to provide defensive artificial intelligence systems ahead of the U.S.’s recent attacks on Iran. Allowing OpenAI’s models in war, as he saw it, was “haphazard.” He described it as a sellout.
He’s not alone in this sentiment. The last few months have seen a growth in frustration over OpenAI’s openness to work with the Trump administration. In and outside the organization, QuitGPT’s mission has grown not just on campus, but all over the Trump-adverse, AI-using world.
I first came across QuitGPT through a flyer left at my door. It was late February, and I was running late to class. “We want your help hosting a QuitGPT party!!!” the flyer said. My roommates and I are used to requests to host — our backyard has seen many social events. Regardless, it’s not every day someone pitches a protest party, especially one calling for a corporate boycott — I was intrigued.
Their flyer made (perhaps intentionally) tall claims — claims that I, as a regularly attentive student with a concerningly high screen time on The New York Times, flagged immediately. “There’s no way that’s true,” I thought to myself.
The flyer claimed, amongst other things, “ChatGPT is Trump’s biggest donor!” and “Their tech enables ICE!”. I showed my roommates, each of us taking turns ridiculing what seemed like a manic cry against a questionable company — sure, nobody was going to defend a company increasingly darkening our job prospects and detaching our understandings of authenticity, but this attack seemed baseless and misguided.
“What about Palantir?” one of my housemates asked, referencing Palantir’s public and proud use of artificial intelligence for military and surveillance purposes. “Why are they just going against ChatGPT?” I had the same question. Why boycott a company for the actions of the Trump administration, especially when the reasons for boycott were so far-fetched?
Their website helped answer some of the far-fetched claims: “OpenAI president Greg Brockman and his wife donated $25 million to MAGA Inc. in 2025 (and CEO Sam Altman donated $1M to Trump’s 2025 Inaugural Fund),” their homepage says. In response to “enabling ICE,” the homepage answers: “ICE’s resume screening tool is powered by OpenAI’s GPT-4.” Fair enough, I thought. Not exactly a smoking gun, gotcha moment against OpenAI. I use and will continue to use an iPhone even though Apple’s former CEO, Tim Cook, has also made Hajj to Trump’s DC — but perhaps some find more value in politically-aligned consumption than I do.
I was still left wondering, why protest like this? Why beat around the bush, protesting by proxy? What does a second-degree protest say about widespread frustrations with the Trump administration, and what does it say about the social permissibility of complacency? This curiosity introduced me to QuitGPT at Berkeley, whose mostly anonymous members I talked to on the condition of anonymity for fear of defamation charges.
Launched earlier this year, the QuitGPT movement began as a social media-spun boycott of OpenAI, citing the company’s proximity to the Trump administration. It has since grown to a grassroots movement, with current and planned chapters at college campuses across the nation, an organizing member told me. The Berkeley chapter, along with a chapter at Stanford, represent the organization within the Bay Area.
QuitGPT at Berkeley made its campus debut in early March with a late-night chalking of Sproul Plaza. “No Mass Surveillance,” they wrote in yellow chalk, contrasting loudly across the gray cement of UC Berkeley’s most traversed area. The messages spread across the ground, lasting for days under the footsteps of hundreds of pedestrians.
The same week, a Reddit post emerged, calling on UC Berkeley students to delete their ChatGPT accounts for similar reasons. “While OpenAI denies any use of its technologies for autonomous weaponry or mass surveillance programs,” the post writes, referencing the then-fresh Pentagon deal, “the order of events is concerning. One can assume that OpenAI has agreed to some unsavory uses of its technologies (in one way or another). Fellow Berkeley students, will you consider canceling your GPT subscriptions?”
A dialogue pursued in the comments, illustrating the struggles not just digital organizing, but of pinning the tail on the Trump administration donkey. “If I had to stop using a product every time a company/institution does anything against my moral compass, I’d have nothing materialistic left in the world and become a monk,” wrote u/No-War-1042.
This perceived lack of agency, one in which young, mostly broke, increasingly unemployed, and exponentially overwhelmed twenty-something-year-olds feel, is widespread. In a way, it’s a form of social nihilism, a disbelief in the power of the individual, and an acclimation to corporate and governmental impunity. And it was this hurdle, a struggle native to Gen Z, that QuitGPT at Berkeley was trying to jump.
The week after the Sproul chalking, I met with QuitGPT at Berkeley’s main organizers over coffee and a muffin (“…atGPT = Trum…” was still visible on the terracotta bricks as I crossed Sproul Plaza to meet them). The organizers, whom I’ll call S and T, explained the organization’s multi-angled approach: Angered by the Trump administration and fearful of the powers that artificial intelligence gave it, S and T saw direct action towards the arms of the administration — OpenAI in this case — as the most efficient form of protest.
They weren’t protesting AI. Independently, they had some personal concerns of the technology, but as an organization, they made their position clear: we’re not against AI, they reiterated. They were simply against OpenAI. To prove that, they pointed towards early marketing slogans, many of which advocated for the use of Claude, Anthropic’s chat model, and direct competitor to ChatGPT. “Cheat With Claude!” said one slogan.
They explained that OpenAI was selling Americans down the river in exchange for short term favor with President Trump. Their goal, they expounded, was to create enough pushback to dissuade OpenAI (and other companies) from following this behavior — essentially, to punish the company for its eager compliance with the Trump administration, and in doing so, to make an example out of them.
To do this, they sought to boycott the use of ChatGPT, hoping to create an upswell of support in calling on the University to cancel their enterprise contracts. Meanwhile, they also sought to dissuade Berkeley students from pursuing future employment with the company. In a way, they wanted to turn OpenAI’s reputation into that of Palantir — a company so commonly disliked that employment there has become a form of campus social taboo.
But, upon reflection, S and T noted the difficulty in organizing and energizing an occasionally unresponsive student base. Between them, I understood their struggle of bringing an atomized generation together, creating a collective effort in a world so hyper individualized. Our conversation reminded me of a piece I had read in The Daily Californian — “‘We didn’t win the war’: How protest at Berkeley fell quiet” — a retrospective look at the living legacy of campus’s Free Speech Movement (FSM). In it, campus freshman Jackson Woodward spoke with veterans of Berkeley’s more radical past in effort to understand Berkeley’s moderate present.
“I’m waiting to hear when the Cal student body will stand up,” FSM Archives Board President Jack Radey was quoted saying. “It’s a very different student body these days.” When asked about the Free Palestine encampments, Radey responded that occupation was not enough — that it was inward facing, self-focused, and had little outreach.
“Tents on Sproul,” Radey says, “did not meaningfully disrupt the machinery of the campus.”
One of the organizers mentioned their participation organizing No Kings rallies in the Bay Area — an interesting parallel. No Kings protests have attracted an older, mainly left-liberal crowd, many of whom are in search of an outlet for their bubbling, vague anger at the Trump administration. QuitGPT on the other hand, is attempting a different feat: organizing a static force. It’s a feat that goes against Newton’s Law of Inertia — pushing an object at rest into a state of motion. And this new challenge called for new forms of organizing. Cue the flyer at my door — QuitGPT at Berkeley was planning parties.
The week after spring break, students returning to Berkeley were met with a new sign decorating the streets and hallways and pinboards of the wider campus area. “FUCK CHATGPT GO RAVE” the posters read in a black serif font on white, 8.5 by 11-inch printer paper.
The flyers mentioned nothing about QuitGPT at Berkeley — their demands, their reasons, or their stances. It mentioned nothing about Trump, or ICE, or the Pentagon. In fact, it barely said ChatGPT, reading from afar “FUCK CHAT GO RAVE,” with GPT in font small enough one might miss it. The flyers, if seen at a pedestrian’s speed, advertised a party like any other. Was this strategic? A Trojan horse of an organizing attempt, recruiting the revolution through bright lights, loud music, and assuredly the effects of a heavy-handed mixed drink? I’m not one to refuse an interesting evening, so I went.
The location was easy enough to find — QuitGPT at Berkeley had hung a large banner with “Let’s POP! the AI Bubble” visible from the street. I introduced myself to the man at the front door, explaining to him that I was working on a piece about QuitGPT at Berkeley, Trump, and alternative forms of protesting. I asked him his thoughts.
“I’m concerned about OpenAI using their technology and selling out to the government for surveillance,” said Rowan George, the gregarious ginger guarding the entrance. Wearing a dark shirt and silver chain, George was the night’s bouncer — a funny role for a protest. George, a third-year Physics major from Richmond, California, wasn’t a member of QuitGPT at Berkeley, but was in support of their goals.
“I’m concerned about the money that is made by OpenAI being directly deposited into Trump’s bank account from the CEO and CFOs and all other people at the top of the chain that donate a lot of money personally to Trump’s campaigns,” George said. He let me in, and I followed the sounds of music up the stairs and into the large, tan, Warring Street house.
The main dance floor was in a large room, painted blue with murals adorning the lower half of the walls, murals of rainbows and faces, paired with the occasional phrase. “FUCK AI” was written next to the door in fresh ink, a likely product of the night. The DJs, who were a rotating cast throughout the evening, were enclosed in a large chain-linked cage that formed a protective barrier between them and the crowd. The crowd, adorning themselves in monochrome stripes and animal prints, ripped leggings and messenger bags, the occasional fur or knee-high boot, filled into the room slowly throughout the evening — dancing, talking, and taking photos.
I had arrived early, in hopes of talking to some of the rave goers before social inertia swept the crowd into motion. Outside, smoking a newly lit cigarette in one hand and drinking from a red plastic cup in another, was Mia Orr, a UC Berkeley sophomore studying art history. Her cup, a rapidly emptying chalice of some pinkish drink, read “DELETE CHATGPT – FUCK ICE,” in white printed letters. Attached to her shirt was a sticker that read “Save Democracy: Cheat with Claude! #quitGPT.”
I asked her what brought her to a rave like this. “It’s really lovely to actually do things that engage with other people, to dance with them and be happy with them,” Orr said. “It’s time I would have just been on my phone.”
Orr reminded me of something George had said as I walked in. I had asked him for his thoughts on raving as a form of protest, to which he corrected me. It’s not about the rave, he explained — the rave doesn’t fight Trump or ICE or ChatGPT or any of the myriad reasons bringing everyone together. “It’s really just connecting with your community, getting off your phone, dancing with your friends,” said George, “which I think is the ultimate community builder and anti-establishment thing you can do — just hanging out and looking out for your neighbors.”
I thought back to The Daily Cal article on the state of Berkeley’s protests. Perhaps it was true — perhaps the revolutionary energy of this campus has changed. Is a rave the descendant of the Free Speech Movement, the anti-apartheid movement, the Free Palestine encampments? Or has campus succumbed to the size of Trump’s power, and minimized their hope and interest in calling for change? Expletives were shot in many directions that night — Trump, OpenAI, and ICE all received custom messages scribbled on the muraled walls. But was that the extent of the resistance? Reveling in anger?
“There’s only so much we can do,” George said. “I don’t think it’s the solution. But I think it’s a start — it starts by changing the things we can in our daily lives.” I watched as he poured another drink into his “FUCK ICE” cup and entered the dance floor to keep “protesting.”
Featured Image: Helios Lung