They are going to implement an AI detector detector detector.

Fun fact, this loop is kinda how one of the generative ML algorithms works. This algorithm is called Generative Adversarial Networks or GAN.

You have a so-called Generator neural network G that generates something (usually images) from random noise and a Discriminator neural network D that can take images (or whatever you're generating) as input and outputs whether this is real or fake (not actually in a binary way, but as a continuous value). D is trained on images from G, which should be classified as fake, and real images from a dataset that should be classified as real. G is trained to generate images from random noise vectors that fool D into thinking they're real. D is, like most neural networks, essentially just a mathematical function so you can just compute how to adjust the generated image to make it appear more real using derivatives.

In the perfect case these 2 networks battle until they reach peak performance. In practice you usually need to do some extra shit to prevent the whole situation from crashing and burning. What often happens, for instance, is that D becomes so good that it doesn't provide any useful feedback anymore. It sees the generated images as 100% fake, meaning there's no longer an obvious way to alter the generated image to make it seem more real.

Sorry for the infodump :3

Spoken like a true AI.

Indeed.

I have an Immich instance running on my home server that backs up my and my wife's photos. It's like an open source Google Photos.

One of its features is an local AI model that recognises faces and tags names on them, as well as doing stuff like recognising when a picture is of a landscape, food, etc.

Likewise, Firefox has a really good offline translation feature that runs locally and is open source.

AI doesn't have to be bad. Big tech and venture capital is just choosing to make it so.

People already mass report to abuse existing AI moderation tools. It's already starting to be accounted for and I can't imagine it so much as slowing down implementing an anti AI rule if I'm being honest.

Nothing is free, If they can sell ads to people because they don't like AI, they will. They're rebooting it with about the same intent as it was originally designed to have.

JPEG artifacts but language

There's literally an app for it called Sora.

Zero tolerance ban still requires a method of detecting AI content in order to enforce said ban. Having such detection system in place would then just as well give people the option to choose for themselves whether they want to see such content or not. Ofcourse such filter isn't 100% accurate but neither is a total ban. Some of that content will always get through.

Prohibition of something easy to make run at home? An interesting choice.

Not really just short form, it's more of a take on video feeds rather than just the limited length quickcontent Vine was famous about.

Obviously the focus is still on short(ish) content format, but I see more and more people transition to longer videos to deliver content. On YT/Facebook most videos I see nowadays are 10min or above.

More tiktok than vine, really

But does it ban AI content?

Bluesky is decentralized, its just decentralized in a different way than Fediverse apps. A lot of fediverse people assume all decentralization would look like ActivityPub, so they just say its fake decentralization rather than learning how it works. (I know because I used to do the same, then I realized how little I understood it from this great blog post, and have since learned a lot more about it.)

There is already alternative infrastructure available (i.e. Blacksky and a variety of other applications hosted using ATProto (you can see a few here: https://bsky.social/about).

You can use any of these apps while maintaining full control of your own data by running your own PDS, or using any community maintained PDS. If you already have an account on the Bluesky PDS, you can migrate it, retaining all of your data. If you dont feel like migrating yet, you can also just export your rotation key, which would allow you to maintain control of your account even in the event that the Bluesky PDS does become evil or something.

Speaking of ATProto, sprk.so is a similar upcoming app, although I think its going to be more similar to TikTok than Vine.

Yes.

It is actually decentralised as far as I can tell, although I would like to point out that bluesky is also decentralised, while bsky.app has dominant market share, nothing stops anyone from hosting their own, completely independent, instance. Some have already done so.

Nostr is part of the fediverse, as is blueksy's ATProto. Neither Mastodon/LEmmy (ActivityPub) nor Bluesky (ATproto) are decentralized. They are poly-centric, and calling Bluesky poly-centric is still a little bit of a fib. NOSTR is truly decentralized. Here on Lemmy if the admin of piefed.world gets mad at the admin of lemmy.ca we couldnt' communicate anymore unless one of us got a new account. On NOSTR if the relay owner no longer wanted my content, you can connect to a different relay that does. (For example)

So "I know it when I see it" rules, rather than anything rigorously defined.

Assuming this gets any traction at all the witch hunts will be rampant.

Need to get a new testflight if on iphone it has an update, but yeah its still mediocre and kinda pointless to post with the compression and lack of users

most people, unfortunately, don’t seem to think when they see the letters ‘A’ and ‘I’… these people probably would burn sage at the sight of the identity matrix lol.

i think you’re probably wasting your breath here but you seem like you might be cool, so if you’re interested in discussing ML at all reach out fs!

Every single one of those I’d put under the second category. It’d be hard to detect but it’s certainly not subjective. It just depends on how it’s written.