The ‘Whisper Strategy’: Unsealed Docs Say Meta Planned to Discredit Its Biggest Critic

中文版 Chinese version

Scope note: this is a purely external read of public court filings and published reporting (New York Post and The Verge, October 27–28, 2026; Fairplay statement, October 27). The meeting notes are described in a filing and summarised by those outlets; Meta had not, in the Post’s reporting, immediately responded to a request for comment. Allegations in litigation are allegations — what is documented here is what the notes are reported to say, not a court finding.

What the documents are reported to say

The sequence matters:

• March 2023: notes from a Meta meeting about “well-being,” later filed in California federal court, record discussion of how to undermine Jonathan Haidt — the NYU social psychologist who had testified to Congress in 2022 about social media and adolescent mental health and had begun publishing his argument on Substack in February 2023. His book The Anxious Generation had not yet been published.
• The notes, as reported, attribute to an executive named Curtiss — likely research chief Curtiss Cobb, per the Post — a “goal” to discredit Haidt, and the view that the company should “focus on the media” to convince reporters Haidt is “not credible.”
• Staff discussed contacting a media figure referred to as “Gellman” to “write a take-down piece” — the Post notes it is not clear who Gellman is.
• Other notes discussed a “whisper strategy” to “turn academics against Haidt” and ways in which “debunking Haidt can happen,” including getting the academic community to disavow him.
• October 23, 2026: the documents were filed publicly / unsealed, per the Post and Fairplay’s timeline. October 27–28: the Post, The Verge and others reported them, as kids-safety trials involving Meta move forward.

Haidt’s response to the Post was one line: he was “not surprised.” Fairplay executive director Josh Golin’s October 27 statement went further: a company that plots to take down a popular academic “is not a company we can trust with our kids’ well-being,” said Meta was more concerned with reputation than making Instagram safer, claimed the documents also show the well-being team tracking interference by in-house counsel, and called on Congress to pass the Kids Online Safety Act by the end of the year.

Separate the two debates — because only one of them is damning

There is a legitimate scientific debate about Haidt’s thesis. Correlation versus causation in teen depression data, effect sizes, cross-country differences, and whether phone bans help are all genuinely contested among researchers, and a company is entitled — like anyone — to fund research, publish rebuttals, and argue the evidence in the open. If the documents showed Meta commissioning a peer-reviewed critique, there would be no story.

What is reported here is different in kind, not degree:

1. The target was credibility, not claims. “Not credible,” “disavow,” “take-down piece” — none of those phrases describes testing a hypothesis. They describe lowering a person’s standing so his evidence stops being heard.
2. The channels were covert by design. A “whisper strategy” aimed at academics and a placed media take-down only work if the audience does not know the source. Open argument discloses its author; this plan, as described, depended on hiding it.
3. It predates the book. In March 2023 there was no bestseller to rebut — there was testimony and a Substack. The sensitivity, as The Verge’s summary puts it, was to Haidt’s potential to sway the cultural narrative ahead of trials. That is litigation-and-reputation strategy conducted inside a team labelled “well-being.”

Haidt himself made the comparison critics will now repeat: after Meta lost a social media addiction case in Los Angeles earlier this year, he wrote that history would judge the industry “as harshly as the tobacco industry.” The tobacco parallel is usually rhetoric. Documents about discrediting a critic are exactly the kind of artifact that made it stick for tobacco — not proof about the underlying health question, but proof about how the company chose to fight it.

The judgment

My read, on the public record alone: even if Haidt’s science is partly wrong, this was the wrong way to be right. A platform that operates products used by teenagers cannot ask for trust on well-being while its own well-being meeting notes, as filed in court, discuss whisper campaigns against the field’s most visible critic. Every future Meta-funded study on teen mental health will now be read through these notes — which is the practical cost, beyond the legal one: the company has made its own legitimate research harder to believe.

The test to watch is response, not rhetoric. Does Meta dispute the notes’ accuracy, contextualise them as a rejected brainstorm, or stay silent? The Post reported no immediate response. Silence, after documents like this, is also a strategy — and readers should price it as one.

Sources

• New York Post, October 27, 2026: unsealed California federal court documents; March 2023 well-being meeting notes; “goal” to discredit Haidt attributed to “Curtiss,” likely research chief Curtiss Cobb; media focus to convince reporters Haidt is “not credible”; “Gellman” take-down piece discussion; “whisper strategy” to turn academics against Haidt; Haidt “not surprised”; Haidt’s tobacco-industry line after the Los Angeles addiction case.
• The Verge, October 28, 2026, via syndicated summaries: Meta considered a campaign to discredit Haidt as public pressure over teen mental health grew and kids-safety trials approached; Haidt testified to Congress in 2022 and began posting his theories to Substack in February 2023, before The Anxious Generation was published.
• Fairplay, statement, October 27, 2026: Josh Golin calling the documents a secret plot, citing a well-being-team spreadsheet tracking in-house counsel interference, and urging Congress to pass the Kids Online Safety Act by year-end.