Social media platforms are increasingly relying on artificial intelligence to moderate content at scale. But that strategy is backfiring, as automated systems remove harmless posts, wrongfully ban users, and reinforce biases. The result is a growing credibility crisis for platforms that claim AI makes moderation faster and more effective.

AI is not enough to protect online communities from AI-generated harm. In fact, many of the moderation tools deployed to fight spam, hate, and disinformation are creating new victims: regular users whose content gets caught in flawed automated enforcement.

Sometimes you have to fight fire with fire. But when it comes to AI slop and hateful content threatening the safety and value of social media platforms, adding more AI can make the problem worse. At its best, social media can be a haven for people who want to

Source: Ars Technica News