AI is here, and already internet is getting flooded with all kinds of AI-generated content. Reddit will require accounts that exhibit "automated or otherwise fishy behavior" to verify that a human runs them. This new policy was stated by Reddit CEO Steve Huffman in a Reddit post, and reported by Ars Technica. The idea is that people managing Reddit "want to make sure that when you're on Reddit, you know when you're talking to a person and when you're not".
It must be said, that human verification will only happen, if Reddit suspects that an account is a bot. This is "rare" and won't apply to "most users". If the account cannot prove that it's human, it "may be restricted".
Reddit will check if an account is run by a human by using third-party tools. It is made clear that these measures will not expose users' true identity, Reddit username, or Reddit activity. These methods currently explored are passkeys, which can help determine that "a human probably did something". Reddit is also looking into third-party biometric services, like World ID, which uses iris-scanning tech. And if nothing else, the last resort might be third-party government ID services, which are already in use in some geographic areas, like the UK.
But there's more to this. Accounts that use bots in permitted ways will get an App label. Reddit has posted information about how developers can get their apps labelled.
Verification process is clearly needed, since Reddit already removes an average of 100,000 accounts per day that "use nefarious bots and post spam". Reddit also plans to make it easier for users to report accounts that they think are bots.
Reddit hasn't confirmed how much of the content on its site is AI-generated.