Does watching Andrew Tate content cause misogyny?

The claim

"Watching Andrew Tate makes boys misogynistic" is one of the most recycled sentences of the post-2022 internet. It appears in school newsletters, campaign reports, parliamentary evidence and dinner-table arguments, usually stated as settled fact. Its mirror image is recycled just as hard inside the manosphere: that the whole thing is a moral panic with no evidence behind it, invented by media outlets that hate successful men. This page tests both versions against the research that actually exists — the algorithm audits, the surveys and the peer-reviewed studies — and is strictly about that research. It is not about the criminal proceedings involving Tate himself, which this site tracks separately and which involve charges he denies: see Is Andrew Tate in jail?

Who says it

The causal version comes from children's-safety and anti-abuse organisations — Women's Aid describes "the significant impact that misogynistic content online can have" on young people's understanding of relationships ^[4] — from school leaders, whose accounts of manosphere tropes arriving in classrooms run through both major UK studies ^[1][5], and from the platforms themselves, whose 2022 bans classed the content as hate ideology: "Misogyny is a hateful ideology that is not tolerated on TikTok," a spokesperson said when the platform removed his accounts. ^[6]

The dismissal version comes from inside the manosphere and from Tate's supporters: the research is described as agenda-driven, the classroom reports as anecdote, and the whole framing as an attack on self-improvement content for young men. Between the two poles sit the researchers, whose published claims are notably more careful than either camp's summary of them. ^[3]

The strongest version

What the evidence shows

The amplification finding — proven

The strongest and most replicated part of the record is about delivery, not effect. The UCL/Kent/ASCL "Safer Scrolling" study built TikTok accounts around archetypes of teenage boys — interests like masculinity, loneliness, self-improvement — and watched what the For You page served over seven days. Initial recommendations matched the stated interests; they then shifted toward anger and blame directed at women. After five days, misogynistic content — objectification, sexual harassment, discrediting women — had quadrupled, from 13% to 56% of recommended videos. ^[1]

The DCU Anti-Bullying Centre study removed even the interest signal. Ten accounts on factory-blank phones, identified only as male, received misogynistic and male-supremacist content within the first 23 minutes across all male accounts, whether or not the account had sought anything like it — with YouTube Shorts serving more toxic material than TikTok, and male "manfluencers" dominating the recommendations. ^[2] Two independent teams, two platforms, one result: a boy does not need to go looking for this content; the recommender brings it to him.

The association finding — real, but not causal

Women's Aid's Influencers and Attitudes research surveyed children, young people and young adults and found that those exposed to misogynistic social media content, including Tate's, were almost five times more likely than the unexposed to view hurting someone physically as acceptable if you say sorry afterwards. ^[4] Internet Matters told a UK parliamentary inquiry that 42% of children aged 9-16 held neutral or favourable views of Tate, that teachers report rising concern about online misogyny shaping boys' behaviour toward girls, and that 56% of young fathers had a favourable impression of him. ^[5] The peer-reviewed strand adds texture: focus groups with London boys aged 13-14 found Tate's tropes operating as "homosocial currencies" — content boys share, joke about and bond over, which the authors argue normalises the misogyny inside the joke. ^[3]

All of this is genuine evidence of a real association. None of it is causal measurement. A survey taken at one point in time cannot distinguish "watching made him think this" from "thinking this made him watch" — both directions produce the same correlation.

Where the causal evidence runs thin

The researchers themselves are more careful than the recycled claim. The UCL team's strongest offline statement rests on interviews and roundtables with school leaders — testimony that tropes are normalising in person — presented alongside, not as, the algorithmic measurement. ^[1]

Harmful views and tropes are now becoming normalised among young people. Online consumption is impacting young people's offline behaviours, as we see these ideologies moving off screens and into school yards.

That is the study's interpretive claim, built on what schools report seeing. It is serious, and it is not the same thing as measured attitude change in viewers — a distinction the recycled versions on both sides erase.

The scale that makes the question matter

Before the platforms removed his accounts in August 2022, videos under the #AndrewTate hashtag had accumulated 14.1 billion views on TikTok; he had 4.7 million Instagram followers and over 760,000 YouTube subscribers. ^[6] Whether the effect per viewing hour is large or small, the exposure side of the equation is enormous — which is why the causal question is worth answering properly rather than asserting. What happened to that reach after the bans is its own contested question: see Does deplatforming work?

Verdict

Partly supported. The claim bundles three assertions, and the evidence grades them differently. That algorithms push this content at boys who never asked for it: proven, twice over, by independent audits. ^[1][2] That exposure is associated with harmful attitudes toward women: documented in surveys and classroom reports, consistently. ^[4][5] That watching causes those attitudes: not yet demonstrated by any study fetched for this page — the audit designs cannot show it, and the survey designs cannot separate cause from selection. The campaign version overstates the record by converting association into causation; the manosphere's "moral panic, no evidence" version misstates it far more, because amplification and association are both real, replicated findings. The honest summary is uncomfortable for both: the delivery system is proven, the association is real, and the causal question — the one everything else turns on — is still open.