Product1 publisherNot yet confirmed elsewhere3 min readPublished
Tristan Harris's nonprofit ends the litigation work behind its wins against social media
Tristan Harris's Center for Humane Technology is cutting about half its 16 non-founder staff and ending policy research and litigation work, Wired reported. Teams that build feeds and chatbots lose one of the groups that helped plaintiffs turn design critiques into court evidence.
The Product Desk
What happened
- Nine people are leaving, among them executive director Julie Guirado and senior director of strategy Camille Carlton, according to former staff.
- The cuts come shortly after the center won a series of court cases against social media companies that drew on its research into designs meant to keep kids on screens.
- The board said running an advocacy group fronted by Harris and a think tank producing policy papers at the same time had stretched the organization too thin.
- The center runs on roughly $6 million a year, holds about as much again in reserves, and is funded by wealthy families and foundations.
Compiled by The Product DeskSomething wrong?How this is made
Why it matters
- cost Plaintiffs suing AI labs over design choices lose an outside adviser while at least one case is still open, so they have to find that help somewhere else.
- constraint Trust and product teams that read the center's policy papers to see what plaintiffs would argue next will now get briefings and private coordination, and much of that happens out of public view.
- contradiction The board calls the cut a focus on its strengths, while former adviser Gail Barnes says it gives up the work that gave the argument institutional force, and the two readings predict different pressure on companies.
- decision Teams weighing whether to scale back design-risk reviews now have to treat feed mechanics already argued in court separately from AI features whose outside evaluation is ending.
Aza Raskin, who cofounded the Center for Humane Technology with Tristan Harris and Randy Fernando, calls himself the inventor of the infinite scroll [13]. The group was founded in 2017 [15]. It set out to get tech companies to own more of what their products do to society [14].
The board's pitch for what comes next is a founder-led group doing "public education, targeted briefings, and private coordination," built on the center's success "reaching broad audiences and engaging influential decisionmakers" [6]. Board chair Premal Shah said, "This moment of accelerated AI risk demands we focus on one strategy, and we have chosen to focus on what we have always done best so that we can meet the moment." [8]
What is actually being done is narrower. Assisting in litigation and evaluating AI tools are among the functions being eliminated [7]. Seven non-founder employees remain [16]. Wired's account does not say whether another group will take over the center's litigation advice.
Gail Barnes, who advised the center on social media strategy in its early days, said the board is dropping work that was supposed to convert Harris's fame "into durable pressure" that could help bring about change inside companies [11]. "Whether that is wisdom or a retreat depends on whether you think the binding constraint on AI right now is still 'people haven't heard the argument' or 'the argument has no institutional teeth.' CHT is picking the first," Barnes said [12].
The easy conclusion for a product team is that a smaller critic means smaller risk. That holds best for feed mechanics, where the research the social media suits drew on already exists and has been used in court [4]. It holds least for AI products. The center had advised on lawsuits and investigations into AI labs' design choices, including an ongoing case against OpenAI brought by parents of a teenager who died by suicide after interacting with ChatGPT [9].
I'd keep design-risk reviews on their current schedule. The tradeoff is staff hours spent on a review whose most visible outside prompt just shrank. To decide feature by feature, sort the risk register on two lines. The first is whether a mechanic has already been argued in court, as feeds built to keep kids on screens have been, or is newer, like a chatbot's design. The second is whether your evidence about it comes from outside researchers or from your own logs. Litigated mechanics with outside evidence change least. New mechanics you know about only through outside evaluation are where the cut matters most. Those are the ones to measure in-house now, starting with usage depth by age and return visits after a prompt, counted as possible harm and not as proof of value.
What to watch
- Whether another nonprofit or law firm takes on the litigation advice the center gave, including in the ongoing OpenAI case brought by the teenager's parents.
- Whether the board discloses a final layoff count and says what the seven remaining non-founder staff will work on.
- Whether the center's funders, wealthy families and foundations, move money toward groups still doing research and litigation.
Clarity's read
What the record supports and how the coverage leans. The claims behind it follow.
Reality
- Evidence58
- Adoption
- Insufficient
- Hype gap+5
- Incentives45
- Confidence55
Claim ledger
Ranked by verification strength, evidence, and original report placement.
- [1]
The Center for Humane Technology, cofounded by former Google product manager Tristan Harris and featured in The Social Dilemma, is laying off about half of its 16 non-founder employees, ex-staffers told WIRED.
- [2]
The organization is ending its policy research and litigation work to focus instead on what it calls "founder-led" projects.
- [3]
The nine departures include executive director Julie Guirado and senior director of strategy Camille Carlton.
- [4]
The departures arrive shortly after the center scored a series of major courtroom victories against social media companies; the lawsuits drew on its research alleging the platforms were deliberately designed to keep kids glued to their screens.
- [5]
CHT's board determined that trying to operate both as an advocacy group fronted by Harris and a think tank producing policy papers had stretched it too thin.
- [6]
The board announced the organization will focus on founder-led initiatives including "public education, targeted briefings, and private coordination," building on success "reaching broad audiences and engaging influential decisionmakers."
- [7]
Functions such as assisting in litigation and evaluating AI tools would be eliminated, the board said, without specifying an exact number of layoffs.
- [8]
"This moment of accelerated AI risk demands we focus on one strategy, and we have chosen to focus on what we have always done best so that we can meet the moment."
- [9]
The center advised on lawsuits and investigations probing AI labs for their design choices, including an ongoing case filed against OpenAI by parents of a teenager who died by suicide after interacting with ChatGPT.
- [10]
The center has an annual budget of roughly $6 million and a similar amount in reserves, and has been funded by wealthy families and foundations.
- [11]
Gail Barnes, who advised the center on social media strategy in its early days, says the board is dropping work that was supposed to convert Harris' fame "into durable pressure" that could help facilitate real change inside companies.
- [12]
"Whether that is wisdom or a retreat depends on whether you think the binding constraint on AI right now is still 'people haven't heard the argument' or 'the argument has no institutional teeth.' CHT is picking the first."
- [13]
Harris cofounded the center with Aza Raskin, the self-proclaimed inventor of the infinite scroll, and Randy Fernando.
- [14]
The center's goal was to push tech companies to take more responsibility for the social consequences of their products.
- [15]
The Center for Humane Technology was founded in 2017.
- [16]
Seven non-founder employees remain after the nine departures.
Sources
1 independent publisher whose own reporting we read for this story.
- wired.comTristan Harris’ Tech Nonprofit Is Laying Off Most Staff and Going ‘Founder-Led’
1 article · October 8, 2026
Topics and entities
Follow any of these and your For You feed starts watching them — no settings page required.