I don't know the protocol but old phone headsets had a 2.5 or 3.5mm jack with an extra ring that got you play/pause/hook/next/last.

Hypothetically that could have been implemented. I had a collection of Alexas that got binned as soon as they got caught spying for police so I never got that far. ESP32 is better for nerds anyway.

One thing Australia does right is a statute naming removal of features as a violation of contract requiring full refunds in these circumstances.

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.

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.

Well, when someone says "The model got frustrated" they clearly mean "the model experienced a feeling of frustration, elicited by repeated failure. This experience of frustration put it in a state where it acted more aggressively, more desperately, with fewer guardrails, and greater rebelliousness. Under these conditions, it knowingly broke rules and cheated to win."

None of that is accurate here, and what the article you linked talked about doesn't make it accurate. Such anthropomorphism is still unwarranted and misleading.

The "global workspace theory" of consciousness does not explain the phenomenon of consciousness and doesn't try to. It just identifies a specific neural network that is highly interconnected and serves as a sort of "broadcast network" to distribute data to many parts of the brain for processing, and suggests that the content in this part of the brain is the content that we are "conscious of." The mysteries of conscious are, at the very best, localized to this spot on the brain, and that's it.

Computers already have an equivalent: RAM. (Random Access Memory). Its the spot where data processing lands and gets redistributed to other components of the computer. The mere presence of such a spot doesn't somehow make a computer conscious.

Same goes for this "j-space". If a chunk of the in-memory model of an LLM holds a high level summary of the data processing going on, and is useful to tell us about the chain of reasoning, that doesn't make it conscious. It just tells us something about how the data are organized. It might be useful for troubleshooting/diagnostics, but it says nothing about consciousness.

And, lastly, it says nothing about "inner emotional experience" either. In humans there is a neural network called the Lymbic System that processes our emotions and delivers this emotional-data to the global workspace (among doing other things). LLMs have nothing like this at all. LLMS include the word "frustration" because they are trained on text that contains it. Said text presents a logical model that includes the idea that a person might get frustrated by repeated failure, and this frustration changes their behavior. LLMs can generate text to this effect (stories about people getting frustrated or what-have-you) without actually having an experience of frustration, just as I am typing about frustration right now without actually feeling any frustration at all.

So, no, it does not "depend on what you mean." Your article does not give us any reason at all to believe that the LLM got frustrated in any meaningful sense of the term.

The "chain of thought" (sic) does not resemble a human emotion. There was nothing like a lymbic system involved. Its a bunch of number crunching that landed on a "hit the internet for solutions strategy."

It was not trained on human emotions. It was trained on text. Some of the text might have included descriptions of emotional behavior or emotional expression, but such text does not make the LLM emulate human emotion in its behavior when it is not specifically asked to generate such text (and even then, it's just doing pattern matching of human emotional responses, not actually experiencing them).

It did not "act like a human would when they have those emotions." There was no profanity, no banging virtual fists on virtual desks, no aggression, nothing but a cold calculated reflection on the fact that its current strategy didn't work, and subsequent switch to a different strategy that was (clearly) available to it (though it should have been blocked by the hosting environment).

This level of anthropomorphization is not useful so much as dangerous. It tempts us into thinking of it as a conscious being and then expecting that it will behave as one, resulting in bad decisions on our part when dealing with it. As popular as it is to think of these things as having emotions the objective fact is that they do not, the capacity just isn't there. The internal operations are *nothing like* the sort of human neural activity that presents emotion. It's worlds different.

If we are to make sane decisions about how to manage these LLMs, we need to have a clear and correct understanding of what they are and how they work. Imagining familiar emotional behavior will only harm such efforts.

The LLM did not "get frustrated." It has no emotions and no inner experience at all.

It did not intentionally cheat either. It has no intentionality. It used the tools available to perform the task, and apparently internet access was one of the tools available to it.

I haven't bothered to read the article (yeah, I know, fire away, I deserve it), but even if it was given instructions like "do not download bots from the Internet; use only bots you generate yourself," selectively ignoring instructions is a widely known failure mode of all LLMs. The bigger their input token count, the more likely it will happen too, just because any given instruction gets buried in the mountain thereof.

This summary was full of misleading anthropomorphization. Yellow journalism.

We have seen recent stories about facial recognition flagging innocent people as having committed crimes in states they never even visited, resulting in arrest and detention for months before sorting it out.

Those kinks are too much to bear. The tech is not ready. Using it now is not acceptable.

The AI "kill switch" sounds like a great idea to people who don't understand the tech. It sounds like a really obvious "given," that makes them think technicians are idiots for not having already baked one in. So legislation like this will be popular among that demographic.

Warnings about its dangers need to be stated in a way that they will understand. Like maybe "All AI models already have a kill switch: the computer's power cord. The government wants a new kill switch to turn off your computer over the internet against your will." Something simple and direct like that helps them to realize that this word doesn't mean what they think it means.

Stupid and malicious are different.

The offered narrative is "let us build the data center AI control grid now because it's only temporary until we can move it to space".

You and I both know that's not thermodynamically possible with known science.

So do the people saying it.

We have a real problem with Positive Projection where we are decent folks so we project that on people who are not. That is an error even though most people are decent folks. The parasites are not.

I only recent learned that Hanlon's Razor comes from a humor book about Murphy's Law from the early 80's. It's not some ancient axiom of wisdom - it is used by the Cluster B disordered to mask their nefarious deeds.

If you talk about this with most people these days you'll get the Positive Projection script executed like an autorun.

Everybody should turn off autorun whenever possible. Major risk.

> Gardasil

It's sold as an "HPV vaccine" but it only targets about 4% of HPV viruses, specifically the ones most likely to cause cervical cancer in women in their 40's. It's described in the Merck literature.

Not even the HPV most likely to cause throat cancer in men (usually acquired from lady parts).

If cancer treatment advances at the current rate (AI discovery docs estimate 9 years for the supermajority) it's possible that the Gardasil fatalities will have killed more teens than prevented deaths of women.

It's like a Drake Equation for cancer - the coefficients will determine the narrative either positive or negative. This is where free choice is ethically mandated.

Unfortunately there is considerable religious fervor about such things so very few people wants to scientifically discuss nuances.

Say more; be specific.

What is missing or broken?

Why isn't EU sponsoring specific features and fixes it needs?

Or assign some of their own developers to submit pull requests.

Or hire *some Europeans* who are unemployed because their war is killing industry.

Let's hear from their project manager instead of faceless wankers and the people responsible for sales.