tekato

Yes, but confidence values are not magic. These values are calculated based on how familiar the current input is to a previous observed input. If the type of input is unfamiliar to the model, what do you think happens? Usually, there will be a category with a high enough confidence score so that it will be chosen as the correct one, while being wrong. Now, assuming you somehow manage to not get a favorable confidence score for any decision. What do you think happens in that case? I never encountered this, but there can only be 3 possible paths: 1) Choose a random value. Not good. 2) Do nothing. Not good. 3) Rerun the model with slightly newer data? Maybe helps, but in the case of driving a car, slightly newer data might be too late.

I love how you quoted all the parts expect the one that mentions where for this to even apply the person have to misuse corporate assets in the first place.

Didn’t bother responding to it because it’s irrelevant. Elon Musk is not being accused of misusing corporate assets, he is being accused of not complying with the DSA.

there is extensive legal precedent in the EU that covers this

Of course, extensive means many. Point to at least 2 cases in which other assets of a company’s owner were threatened by the EU government due to DSA violations. I know there are none, so I’ll make it easier. Name 2 cases for which the EU went after the owner’s other assets because the fine to the company wasn’t enough, whatever the fine might originate from.

So a company provides infinite protection?

No. Never even suggested that.

"I didn't murder that man, the company did."

"The company paid individual X to murder them, not me."

Ridiculous stretch. I have stated multiple times, as has the law, the company provides financial protections, did I ever say anything else? Of course if you are involved in the murder of somebody you should be prosecuted. If you can’t argue in good faith don’t bother responding.

if your name is Elon Musk and you own companies X, Y, & Z, and you perform actions A, B, C, you I'll be fined in this exact way?

No. You can clearly state in the law that “if your company is found to violate a consumer protection law, your other assets will be in jeopardy if we can’t figure out a way to fine your company”. I wonder what would be the consequences of explicitly informing companies of the consequences of their actions.

And do you know who agreed to these rules? Elon Musk. He chose to do business in the EU. He agreed to their rules.

Yes he did. He agreed to the rule that states the company will be fined up to 6% of the yearly income, not whatever this is.

It can't be irrelevant as it's the primary factor in deciding if the fine will even be brought.

Is DSA the only way for your company to get fined? If the answer is no, then yes, it’s irrelevant. Because while your company may not be eligible for a DSA fine, there are countless different situations which could leave you in the same spot.

Personal liability applies to many actions under the law

Yes, but none of those actions involve what is happening here. The DSA clearly states that the company may be fined up to 6% of its yearly revenue.

your scaremongering of small family business becoming some governments targets are unfounded.

What scaremongering? This is a valid concern. If Elon Musk’s rights as a company owner can be violated, who says yours can’t?

Actions you don’t like aren’t corrupt. Actions Elon doesn’t like aren’t corrupt.

Can you point to where I said or implied the opposite?

In fact arguably Elon is the corrupt individual in this case

I never said Elon was in the right. I have stated that his company must be punished for the decisions it made, not the owner as per the protections guaranteed by the law to company owners.

the EU is simply applying the law to the corrupt individual.

You mean the law that violates your rights as the head of a corporation, which is supposed to protect your assets from those of the company? Cool.

Right, which is why that marvelous confidence value got somebody ran over.

This is why you can’t have an AI make decisions on activities that could kill someone. AI models can’t say “I don’t know”, every input is forced to be classified as something they’ve seen before, effectively hallucinating when the input is unknown.

In an ideal world the headline would be “Google kills Chrome by preventing users from blocking ads”.

Imagine buying a politician’s cryptocurrency

FSR is an AMD technology and it’s not AMD exclusive. So one doesn’t imply the other.

You are going for the assets of the owner, which are unrelated to the company in question. Effectively devaluating companies that are not the infringing one, which directly affect the person’s net worth. This is no different than the government straight up taking money from your investments or 401K.

Whether it’s subject to DSA or not is irrelevant. The fact is that a company has to pay a fine to the government, whatever the infraction might be, and the government is trying to force something else, not the company in question, to pay the fine.