The FTC is looking into how Reddit licenses data before its IPO

Reddit is planning on going public. But first, the Federal Trade Commission has to check on some stuff.

The social media company revealed that on March 14, the FTC sent them a letter declaring that the commission would conduct a "non-public inquiry" regarding Reddit's licensing of user-generated data for training third-party AI models, per CNBC. Data licensing is a young business for Reddit, which relies mostly on advertising revenue at this point.

Reddit's plan at the moment is to go public on March 21. It is unclear what kind of effect, if any, this FTC investigation will have on those efforts.

As you may or may not be aware, many generative AI tools like ChatGPT, Sora, and Midjourney require a great deal of data to train on. The easiest, and most accessible source for huge data sets like the ones these tools require is the internet. Some companies like Tumblr owner Automattic have been willing to strike deals with AI companies to license data for training. That's what Reddit is trying to do here.

Arguably the most prominent third-party AI model that Reddit has partnered with for training is Google's, which was announced last month. We'll see if it winds up haunting Reddit at some point down the road.