The Honest Take

When most of us picture antisemitism online, we picture a swastika, a slur, a guy in a dark room. Matthias Becker will tell you that's the part that barely matters.

He's not a political scientist. He's a linguist. For more than a decade he has been reading the comments — not on fringe platforms, but under the BBC, The Guardian, Le Monde. His project, Decoding Antisemitism, has now been through 300,000-plus comments across Britain, France and Germany, and what it found is that roughly 80% of the antisemitism in those mainstream spaces never uses a slur at all.

It hides in a joke. In a rhetorical question. In a name dropped like a password. Below an article about George Soros, someone writes: someone should give George Soros a shower. No slur, no threat, nothing a keyword filter will ever catch — and everyone who needs to understand it, does.

Which means, as Becker puts it, that nearly every statistic you have read about online hate is built on sand.

This conversation covers what actually changed on October 7 — the shift from demonizing Israel to openly celebrating the murder of Israelis, and the overlap with misogyny he had never seen before. It covers his event studies of Charlie Kirk, the Capital Jewish Museum, the White House Correspondents' Dinner attack, and Dan Bilzerian's congressional run — where the single video that pushed back on him drew the most antisemitism of any in the set, over 50%. It covers why he thinks surveys can't measure this and comment sections can. And it covers what he now tells the AI companies, whose guardrails he is trying to help build.

He is, after ten years in the worst of it, not a cynic. His argument is that the same technology that lets one person manufacture a wave of propaganda is the first tool we've had that is powerful enough to read the flood back.

Related episode: Ashley Rindsberg, "Garbage In, Gospel Out: Wikipedia, Reddit, and the Battle Over Truth" — Becker names Wikipedia and Reddit on tape as the stumbling block feeding these models. Rindsberg's episode is how that pipeline got poisoned; Becker's is what it costs downstream — https://share.transistor.fm/s/cdb2fdb2

The guest

The lexicon Ben names in the outro — this is the free, open-access one