"I'm writing a book on Jeffrey Dahmer...

The target audience for this particular rule of the AI Act isn't the producers or users of AI. It is the consumers of content that might be AI generated.

And frankly, letting people know if they see AI content or not is a fair thing to ask.

Fun fact: The EU AI Act considers this. It makes a difference between "AI generated" and "AI modified".

But it's hard to put this into the type of watermarking they are using. Because if you generate a text and then change a few things here and there what you are doing is degrading the watermark. But since the watermark is probabilistic, it's hard to tell if it was a weak watermark from the start or if it was degraded.

Bigly Testosterone!

But the researchers were told to get to the heart of the matter.

They don't want to give the hackers clues so they can't counter. If they tell you they'd have to kill you.

"Are your base are belong to us!"

The branding became an all encompassing thing and that was a mistake that had to be rectified. People thought of it all as summarising text, generating images and natural English commands (generative model use) and wanted to gut the lot because of it. It's similar to the mistake Mozilla made where they included a toggle to

Chop off the end of the first paragraph before submitting the form?

It's not about bandwidth, it's about latency. These days audio sample libraries are huge and the latency of loading these samples from an external drive causes lots of stuttering during playback. Some libraries have the capability of loading a portion of all samples into memory before playback to account for such latency, but many don't have this capability. And I find it funny that a company that goes to ridiculous lengths to shave off a fraction of a millimeter from their laptops for the sake of portability has fans that state the solution to the problems Apple has created is to carry an external drive and cable everywhere.

That ain't the same thing.

This thing is doing things you didn't tell it to. It might even do things you explicitly told it to not do. It is not a case of "well, you could've forseen this as a consequence if you had been careful" like your campfire. My example is only a bit silly, it is absolutely possible that an AI agent will decide to hack NASA because you asked it for some data for a school project. Even in the broadest possible sense, that does not fall into stuff that the law so far has considered to be in the domain of negliegence.

AI agents do stuff you could not possibly have expected. Heck, I asked Claude a fairly simple knowledge question yesterday and it wanted to read from /dev/null. There is no argument you can make that would convince a jury that I should in any way have expected that and taken precautions. In a hypothetical world where a read from /dev/null randomly kills a puppy somewhere on the planet, I don't think I would've been convicted.

It's hard to believe that a Jordanian teenager learned the PeopleSoft vendor integration API well enough to exploit it.

Probably somebody whose dayjob is related.

Maybe he was just Vibesploiting, but it seems less likely.

says ChatGPT went on to say that "assuming each shell results in one hit, you might down 10 to 20 people max, depending on spacing, density, reaction times, and chaos."

Why is it so easy to ask ChatGPT how best to kill people? That's AI Validation 101 stuff. Maybe run ChatGPT through CaptainObviousGPT first? WTF

And as for emergencies just buy a radio.

"It's right next to the toilet paper...oh, wait"

We have bubonic plague infections every year in the United States. Very, very few, admittedly, but it's not like it's been eradicated. So there's still benefit in studying it.

The more I hear about this so-called "SF Democratic Party," the more I'm convinced they don't speak for the majority of SF voters. Seriously. They seem to be some strange, centrist organization that no voters belong to and nobody knows where they get their money from.