> How can an algorithm determine whether content is political or not? For starters, cross-referencing it with popular propaganda talking points. Smear campaign

> How can an algorithm determine whether content is political or not? For starters, cross-referencing it with popular propaganda talking points. Smear campaigns are a dead giveaway, and adding two or three topics should be enough to cover the majority of propaganda posts. I'd argue that just having an effective anti-bot policy is enough to render this a non-issue, but apparently Elon Musk prefers Twitter to be a Kremlin bot sandbox.