One of the most powerful arguments against hate speech is that it is dangerous: it increases the likelihood that listeners will engage in violence and discrimination against targeted groups. Banning hate speech, then, is justified as a strategy for crime prevention. Among the most important objections to this argument is that it is incompatible with a view of listeners as autonomous. If we regard moral agents as capable of making up their own minds about what to believe and do, as we surely should, we cannot countenance efforts by the state to censor the principles and proposals to which they are exposed. This paper aims to evaluate whether this autonomy objection to bans on dangerous speech is persuasive. If hateful speech should remain legal because listeners can make up their own minds about whether to embrace and act on its sinister message, why shouldn’t ordinary incitement to crime (such as exhorting someone to commit a murder) be legal for the same reason? I argue that defenders of the autonomy objection to hate speech bans face the following choice. Either they should embrace the legalization of incitement to crime—thereby running afoul of both established Western legal practice and widespread moral conviction—or else they should renounce their autonomy-based opposition to bans on hate speech. The paper defends the latter position.