Nothing in law is simple. Legislators pass a law and after it passed that meat grinder it eventually gets brought up in courts where a second sausage is made by evaluating whose rights take precedence over others. And if we imagine the current US Supreme Court they will not ban deepfakes completely because it could limit the first amendment. And they would probably find in favor of the sad smelly asshole in his basement who made deepfake porn that was never meant to be public. So the creation of deepfakes will not be limited by law. The system can only react after the creation. By which point the damage is already done for the celebrity whose likeness was abused.

Right now it might be possible to get the companies to limit the models that can generate this stuff. But soon enough, maybe in a few years, it will be possible to train you own model on your terms that will run locally and if you’re savvy enough to set that up you’ll be savvy enough to sidestep any restrictions there as well.

Don’t impersonate without enthusiastic consent will not survive the tour up the legal system.

Yeah but it isn’t here.

All this so-called-AI generated policy meat still needs to go through the sausage grinder in both houses, filled with humans. So I think the scandal wouldn’t be as great.

The danger depends more on who you are. If you are in a position where deepfakes could be made to undermine you in your life, there is a higher danger. If you work a desk job in an accounting firm, that risk is much lower.

Deleting all your pictures from the internet is a fig leaf. How many pictures exist in other people’s photorolls that you are in? And even if you trusted all of them implicitly, how well do they do their security?

I think at this point in time, deepfakes are ultimately identifiable. By which I don’t mean anybody can tell immediately that it is one. But on repeated viewing enough people would get suspicious and when somebody analyzes the 1s and 0s it can be made certain. Society needs to adapt a delayed response tactic. This will take time but eventually we will look at deepfakes with the same skeptical eye we developed for photoshopped images. We are in the period of adjustment lag so the jeopardy is higher today than it will be a couple of years from now.

The biggest danger is for ladies because sexualized deepfakes are not only appalling but the legal system is also lagging to catch up in many places. As soon as a believable deepfake of a famous man makes the rounds, the laws are going to be tightened ASAP though.

There is also a legal battle that needs to be fought. Can I hypothetically make a deepfake sex video of my favorite female celebrity just for my own enjoyment? I could paint her oil on canvas as long as I kept it at my house. I could write fan fic and might even get away with posting that online. Could I not make this movie just for me? And if I protected my computer in a reasonable way, can I be held accountable if some other asshole leaked it onto the internet? We’ll have the answers in 10-15 years.

I don’t think you can make all deepfakes illegal so we’ll have you find a way to live with the threat.

If you want to get picky, Xwitter didn’t enshitify as laid out as a concept by Cory Doctorow. The best example is probably Amazon which went from being insanely user friendly to lock in users, to supplier-friendly and increasingly less so for users, until it had squeezed and shafted both groups. That’s enshitification and it doesn’t apply to Xwitter. They had problems to make money before a certain somebody bought it. They’ve been bleeding users since the eventually Nazi saluting manbaby bought it, who then wanted to sue advertisers who refused to buy ads on his service. There was no user lock-in and then a supplier lock-in. There was just shit. All their current problems are man made. By one specific man.

I would argue you cannot enshitify a service that was already shit. At this point this is more of a conshitidation.

Oh, man, this will affect tens of people.

I’m concerned about the resources it gobbles up but I’m not skynet scared.

The effectiveness and the benefits are overhyped by the people with a financial interest in it. We call it AI but it’s only one for two there in that it is artificial. It’s a technology in its infancy. And it hasn’t found a use case that will guarantee an ROI. And the companies involved here are running out of runway.

Cannibalism hasn’t been proven yet. But even without that, rich people not getting punished isn’t mildly infuriating. We are way past that.

And the world was already shit long before we learned about any of this. Because it was possible to get away with that shit before we found out and even still after we learned a little about it.

Billionaires aren’t the majority. Turn your morose mood into a political movement that will hold them to account.

Would the order be different in an imported fridge?

The sliding scales of inviting all the people you want to have there and avoiding people getting pissed off if they don’t get an invite (or similar political reasons) are only limited by the financial means available.

100 is a relatively easy target to reach for most people. Family and friends and their +1’s and children gets you there pretty quick.

If Ockham’s Razor points to the simplest possible reason of anything to be the most likely one, Drumpf’s Razor, which I have just made up, holds that the dumbest possible explanation is probably true when the incumbent US head of state is involved. These are posts by an undereducated narcissist; the choice of targets is at best accidental.

The prompt probably included “destroy Iran” and whichever model they used was like gotcha, destroying Ireland it is.

I don’t have a GitHub account. You’re gonna have you sign up for one yourself. The links to their accounts are on the website I linked to above. Just click on the avatars.

https://www.fossify.org/ only lists reddit and Telegram of all services. You may be able to get in touch with some of the contributors via GitHub. And you could see if some of them recycle their user names on Mastodon.

Don’t forget the klick. Most of them are not buying that either.

The people in all the countries that have no problem counting off another dozen past twelve don’t always do that though. If you meet your friend at 15:00 most people will revert to “at 3” in their language. And they might “go to bed at 11.” Economy of language and context clues. So colloquially the am/pm crowd and the 24h folks aren’t far apart at all.

And any person claiming that it’s too difficult to add or subtract twelve from at maximum a low two-digit integer ought to have their passport revoked.

There are also canals that aren’t this shallow. As summer lasts on average five minutes, anybody falling into one is more likely to wear thick layers of clothes, which is a problem when they’re getting wet. Most British people are also probably under the influence when they fall into a canal. Or children. And you can drown in a puddle.

I can think of two confederacies that have flags. And you haven’t specified the feed.

Which one? And that applies both to the confederacy and the feed here.

To be fair, if anything on the internet has too much traffic, it becomes unavailable. That strategy is at the heart of (D)DOS attacks.