Unfortunately advertising doesn’t work on the majority of their users who are bots. 🤷

When a company takes on shareholders, whatever goals, mission, or ethos they had is erased. They now exist as a vehicle to make as much money as possible at literally any cost. That’s it. Was nice while it lasted.

Have they tried not using it? 🤦

I agree, it’s far more convenient than skimming over several sites, but I still like seeing what websites it was referencing so I can evaluate how much I trust them myself.

Kagi’s FastGPT. It’s handy for quick answers to questions I’d normally punch in a search engine with the same ability to vet the sources.

I’ve used an LLM that provides references for most things it says, and it really ruined a lot of the magic when I saw the answer was basically copied verbatim from those sources with a little rewording to mash it together. I can’t imagine trusting an LLM that doesn’t do this now.

The reason is that Elon Musk got caught liking questionable content through the likes page.

The big record labels are shareholders in Spotify so they’re happy to get less money in streaming royalties because that’s the part they have to share with artists, but the value of their shares they get to keep all for themselves.

https://www.rollingstone.com/pro/news/who-really-owns-spotify-955388/

I bet the AI was tuned to select ads that maximize both profit and engagement for Meta over maximizing either profit or engagement for the advertiser. Totally working “as intended”.

deleted by creator

Hahah but really AI is already being used to amplify and exploit all the problems of social media to new levels. It was nice while it lasted, but we can’t stuff this all back in Pandora’s box.

Probably all those throwaway accounts that people create to post comments that they don’t want attached to themselves in any way. I doubt many people took enough precautions to prevent Reddit from identifying them as alternate accounts though.

Everyone: Don’t say anything sensitive or personal to an AI because it could end up in training data!

Microsoft: We’re making it easier to feed everything you do on your computer to an AI from notepad to your desktop!

[thisisfine.jpg]

I just made the switch and Steam with Proton has been really smooth, they’ve made a lot of progress to make it easy since the Steam Deck has come out. I don’t play any online competitive games that use anti-cheat though.

Fair, though this is also where the double-edge sword of discoverability steps in too. Many people complain about the lack of it on decentralized systems, but centralized systems have a nice catalog of users for bots to message with little effort.

I’ll admit that lack of discoverability isn’t a perfect solution since there are other ways for spammers to discover users. E-mail is a great example of a large, long running, decentralized system that has increasingly suffered from spam since its inception due to mass data collection of addresses. However if you’re really careful about who you share your address with, it’s possible to still avoid most of it. I give out unique e-mail address to companies and spam tends to only come in on a few, often because they were breached or are otherwise “leaky” about their user’s data. Dropbox is by far the worst offender.

I’ve seen pictures of rooms with walls full of Android cell phones on shelves all hooked up by USB for power and remote control. They can load apps, register accounts, and interact with content inside the app while appearing as legitimate mobile users. That’s why moves like Reddit restricting API access only hurt legitimate users and lazy bot farms, cause the hardcore bot farms have been using the official app on real phones all along.

I’ve been using Mastodon and it’s a pleasant change of pace. I’ve heard of some spam happening there but I think responsive admins and the lack of algorithmic feeds really reduces their reach.

Oh actually it’s worse than that. There are online companies that offer online SMS services that can receive messages from real phone numbers by essentially telling your carrier you want text messages forwarded to them. Obviously they usually make you prove that you own the number before requesting forwarding, but there’s ways around that. I’ve known several people who’ve had their online accounts broken in to because someone hijacked their phone number’s SMS in order to perform password resets or bypass 2FA.

While some may see this as good for Bluesky, I bet this is the floodgates opening to bots and algorithmically boosted harmful content. Good luck everyone on there!

Well if Meta is the “industry leader” of tools designed to prevent this yet it’s still happening at a large scale, then he’s basically admitting that there is no way the industry can solve this. I hope they get legislated into the ground.