It's still a bummer that Reddit had to kill all public data access completely in the process of locking down their data supposedly to prevent AI from scraping it. (but in reality, they did it in order to facilitate these deals)

The PushShift datasets were very useful for giving demos on how big data can be analyzed with more intuitive and interesting results. Here's an older blog post of mine analyzing Reddit data from 2018: https://minimaxir.com/2018/09/modeling-link-aggregators/

I would not be in data science/machine learning now if it weren't for Reddit data.

Who the hell paid 60M for their garbage data? This can't be real. Also another reason why scraping content to use for models has to be free for everyone or incumbents are gonna pull stuff like this.

After the API exodus most of the subs I like were shadows of themselves.

This is probably a good move for the Reddit owners to get some revenue for a site rapidly becoming as relevant as Slashdot.

When am I getting my check from reddit for my content?

AI content licensing for Reddit? This is going to go down about as well as a lead balloon.

Like, if AI video and art are controversial on there, AI use of users' content is going to be even more controversial to say the least. Perhaps enough to cause another protest or outrage to break out.

This feels like the silliest decision Reddit corporate could make with an IPO.

1. A single cancelable $60M per year deal

2. ????

3. $5B IPO price target

4. Profit

So, what's my cut for years of commenting?

This sounds stupid. And that lofty $5B target. Surely Meta don't want Reddit threads at this point? Then there's only one outlier within social media I can think of that is as silly as this and wants the kind of content and tone of Reddit. It's about X and Grok, isn't it?