Well, Digg back then was better than Reddit is now, so I guess that’s something.

Looks like it was a practice run. It’s relatively easy to take out webservers with a standard DDOS attack. This is considerably more sophisticated, and I think they were testing it on gaming networks in prep for a larger attack on financial and/or government IT infrastructure.

Someone’s been munching Elon’s Special K for breakfast. New marketing tagline just dropped tho:

Brave browser, because you’d have to be brave to use a browser built by a Nazi

Well, Reddit’s approach towards AI and auto-mod has already killed most of the interesting discussion on that site. It’s one of the reason I moved to the Fediverse.

At the same time, I was around in the Fediverse during the CSAM attacks, and I’ve run online discussion sites and forums, so I’m well aware of the challenges of moderation, especially given the wave of AI chat-bots and spam constantly attempting to infiltrate open discussion sites.

And I’ve worked with AI a great deal (go check out Jan - open source, runs on local machine if you’re interested), and there’s no chance in hell it’s anywhere near ready to take on the role of moderator.

See, Reddit’s biggest strength is its biggest weakness = the army of unpaid mods that have committed untold numbers of hours towards improving the site’s content. What Reddit found out during the API debacle was that because the mods weren’t paid, Reddit had no recourse to control them aside from “firing” them. The net result was a massive loss of editorial talent, and the site’s content quality plunged as a result.

Because although the role of a mod is different in that they can’t (or shouldn’t) edit user content, they are still gatekeepers the way junior editors would be in a print publishing organization.

But here’s the thing - there’s a reason you pay editors. Because they ensure the content of the organization is of high caliber, which is why advertisers want to pay you to run their ads.

Reddit thinks it can skip this step. Instead of doing the obvious thing = pay the mods to be professionals - they think that they can solve the problem with AI much more cheaply. But AI won’t do anything to encourage people to post.

What encourages people to post is that other people will see and comment, that real humans will engage with their content. All it takes is the automod telling you a few times that your comment was banned for X inexplicable reason and you stop wanting to post. After all, why waste your time creating unpaid content for a machine to reject it?

If Reddit goes the way of AI moderation, they’ll need to start paying their content creators. If they want to use unpaid content from an open discussion forum, they need to start paying their moderators.

But here’s the thing. Reddit CAN’T pay. They’ve been surfing off of VC investment for two decades and have NEVER turned a profit, because despite their dominance of the space, they kept trying to monetize it without paying people for contributing to it… and honestly, they’ve done a piss poor job at every point in their development since “New Reddit” came online.

This is why they sold your data to Google for AI. And its why their content has gone to crap, and why you’re all reading this on the Fediverse.

Counterpoint - if you must rely on AI, you have to constantly exercise your critical thinking skills to parse through all its bullshit, or AI will eventually Darwin your ass when it tells you that bleach and ammonia make a lemon cleanser to die for.