In early February, animal welfare advocates and AI researchers gathered in stocking feet at Mox, a scrappy, shoes-free coworking space in San Francisco. Yellow and red canopies billowed overhead, Persian rugs blanketed the floor, and mosaic lamps glowed beside potted plants.
In the common area, a wildlife advocate spoke passionately to a crowd lounging in beanbags about a form of rodent birth control that could manage rat populations without poison. In the “Crustacean Room,” a dozen people sat in a circle, debating whether the sentience of insects could tell us anything about the inner lives of chatbots. In front of the “Bovine Room” stood a bookshelf stacked with copies of Eliezer Yudkowsky’s If Anyone Builds It, Everyone Dies, a manifesto arguing that AI could wipe out humanity.
The event was hosted by Sentient Futures, an organization that believes the future of animal welfare will depend on AI. Like many Bay Area denizens, the attendees were decidedly “AGI-pilled”—they believe that artificial general intelligence, powerful AI that can compete with humans on most cognitive tasks, is on the horizon. If that’s true, they reason, then AI will likely prove key to solving society’s thorniest problems—including animal suffering.
To be clear, experts still fiercely debate whether today’s AI systems will ever achieve human- or superhuman-level intelligence, and it’s not clear what will happen if they do. But some conference attendees envision a possible future in which it is AI systems, and not humans, who call the shots. Eventually, they think, the welfare of animals could hinge on whether we’ve trained AI systems to value animal lives.
“AI is going to be very transformative, and it’s going to pretty much flip the game board,” said Constance Li, founder of Sentient Futures. “If you think that AI will make the majority of decisions, then it matters how they value animals and other sentient beings”—those that can feel and, therefore, suffer.
Like Li, many summit attendees have been committed to animal welfare since long before AI came into the picture. But they’re not the types to donate a hundred bucks to an animal shelter. Instead of focusing on local actions, they prioritize larger-scale solutions, such as reducing factory farming by promoting cultivated meat, which is grown in a lab from animal cells.
The Bay Area animal welfare movement is closely linked to effective altruism, a philanthropic movement committed to maximizing the amount of good one does in the world—indeed, many conference attendees work for organizations funded by effective altruists. That philosophy might sound great on paper, but “maximizing good” is a tricky puzzle that might not admit a clear solution. The movement has been widely criticized for some of its conclusions, such as promoting working in exploitative industries to maximize charitable donations and ignoring present-day harms in favor of issues that could cause suffering for a large number of people who haven’t been born yet. Critics also argue that effective altruists neglect the importance of systemic issues such as racism and economic exploitation and overlook the insights that marginalized communities might have into the best ways to improve their own lives.
When it comes to animal welfare, this exactingly utilitarian approach can lead to some strange conclusions. For example, some effective altruists say it makes sense to commit significant resources to improving the welfare of insects and shrimp because they exist in such staggering numbers, even though they may not have much individual capacity for suffering.
Now the movement is sorting out how AI fits in. At the summit, Jasmine Brazilek, cofounder of a nonprofit called Compassion in Machine Learning, opened her sticker-stamped laptop to pull up a benchmark she devised to measure how LLMs reason about animal welfare. A cloud security engineer turned animal advocate, she’d flown in from La Paz, Mexico, where she runs her nonprofit with a handful of volunteers and a shoestring budget.
Brazilek urged the AI researchers in the room to train their models with synthetic documents that reflect concern for animal welfare. “Hopefully, future superintelligent systems consider nonhuman interest, and there is a world where AI amplifies the best of human values and not the worst,” she said.
The power of the purse
The technologically inclined side of the animal welfare movement has faced some major setbacks in recent years. Dreams of transitioning people away from a diet dependent on factory farming have been dampened by developments such as the decimation of the plant-based-meat company Beyond Meat’s stock price and the passage of laws banning cultivated meat in several US states.
AI has injected a shot of optimism. Like much of Silicon Valley, many attendees at the summit subscribe to the idea that AI might dramatically increase their productivity—though their goal is not to maximize their seed round but, rather, to prevent as much animal suffering as possible. Some brainstormed how to use Claude Code and custom agents to handle the coding and administrative tasks in their advocacy work. Others pitched the idea of developing new, cheaper methods for cultivating meat using scientific AI tools such as AlphaFold, which aids in molecular biology research by predicting the three-dimensional structures of proteins.
But the real talk of the event was a flood of funding that advocates expect will soon be committed to animal welfare charities—not by individual megadonors, but by AI lab employees.
Much of the funding for the farm animal welfare movement, which includes nonprofits advocating for improved conditions on farms, promoting veganism, and endorsing cultivated meat, comes from people in the tech industry, says Lewis Bollard, the managing director of the farm animal welfare fund at Coefficient Giving, a philanthropic funder that used to be called Open Philanthropy. Coefficient Giving is backed by Facebook cofounder Dustin Moskovitz and his wife, Cari Tuna, who are among a handful of Silicon Valley billionaires who embrace effective altruism
“This has just been an area that was completely neglected by traditional philanthropies,” such as the Gates Foundation and the Ford Foundation, Bollard says. “It’s primarily been people in tech who have been open to [it].”
The next generation of big donors, Bollard expects, will be AI researchers—particularly those who work at Anthropic, the AI lab behind the chatbot Claude. Anthropic’s founding team also has connections to the effective altruism movement, and the company has a generous donation matching program. In February, Anthropic’s valuation reached $380 billion and it gave employees the option to cash in on their equity, so some of that money could soon be flowing into charitable coffers.
The prospect of new funding sustained a constant buzz of conversation at the summit. Animal welfare advocates huddled in the “Arthropod Room” and scrawled big dollar figures and catchy acronyms for projects on a whiteboard. One person pitched a $100 million animal super PAC that would place staffers with Congress members and lobby for animal welfare legislation. Some wanted to start a media company that creates AI-generated content on TikTok promoting veganism. Others spoke about placing animal advocates inside AI labs.
“The amount of new funding does give us more confidence to be bolder about things,” said Aaron Boddy, cofounder of the Shrimp Welfare Project, an organization that aims to reduce the suffering of farmed shrimp through humane slaughter, among other initiatives.
The question of AI welfare
But animal welfare was only half the focus of the Sentient Futures summit. Some attendees probed far headier territory. They took seriously the controversial idea that AI systems might one day develop the capacity to feel and therefore suffer, and they worry that this future AI suffering, if ignored, could constitute a moral catastrophe.
AI suffering is a tricky research problem, not least because scientists don’t yet have a solid grip on why humans and other animals are sentient. But at the summit, a niche cadre of philosophers, largely funded by the effective altruism movement, and a handful of freewheeling academics grappled with the question. Some presented their research on using LLMs to evaluate whether other LLMs might be sentient. On Debate Night, attendees argued about whether we should ironically call sentient AI systems “clankers,” a derogatory term for robots from the film Star Wars, asking if the robot slur could shape how we treat a new kind of mind.
“It doesn’t matter if it’s a cow or a pig or an AI, as long as they have the capacity to feel happiness or suffering,” says Li.
In some ways, bringing AI sentience into an animal welfare conference isn’t as strange a move as it might seem. Researchers who work on machine sentience often draw on theories and approaches pioneered in the study of animal sentience, and if you accept that invertebrates likely feel pain and believe that AI systems might soon achieve superhuman intelligence, entertaining the possibility that those systems might also suffer may not be much of a leap.
“Animal welfare advocates are used to going against the grain,” says Derek Shiller, an AI consciousness researcher at the think tank Rethink Priorities, who was once a web developer at the animal advocacy nonprofit Humane League. “They’re more open to being concerned about AI welfare, even though other people think it’s silly.”
But outside the niche Bay Area circle, caring about the possibility of AI sentience is a harder sell. Li says she faced pushback from other animal welfare advocates when, inspired by a conference on AI sentience she attended in 2023, she rebranded her farm animal welfare advocacy organization as Sentient Futures last year. “Many people were extremely confident that AIs would never become sentient and [argued that] by investing any energy or money into AI welfare, we’re just burning money and throwing it away,” she says.
Matt Dominguez, executive director of Compassion in World Farming, echoed the concern. “I would hate to see people pulling money out of farm animal welfare or animal welfare and moving it into something that is hypothetical at this particular moment,” he says.
Still, Dominguez, who started partnering with the Shrimp Welfare Project after learning about invertebrate suffering, believes compassion is expansive. “When we get someone to care about one of those things, it creates capacity for their circle of compassion to grow to include others,” he says.