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.

I'm not saying it is a *good idea* for a community to sell out like that, I'm just saying a majority likely would take the deal.

Finding a single place willing to put up with a datacenter is as much success as deploying thousands of these hypothetical space datacenters.

For the cost of a *single* SpaceX launch, they could basically bribe an entire small town's population to be happy with them. $50K a person and they'll *probably* pass public comment.

That would affect the buyer's rights with respect to the seller, not not his rights with respect to the gaming company.

I wonder whether this is legal. Riot has the right to ban players who cheat, but maybe not the right to misidentify them based on the hardware they are using. In this case, since the game is free, there may be no contract of which this is a breach. In the case of a paid game, it seems that it would be a breach of contract.

You don't get it.

Autonomous AI agents are the perfect no-responsibility card. The AI companies? They trained a general-purpose model and never gave it any explicit instructions to do hacking. The AI services? They are just middlemen and have no control over what people tell the agents to do. The people prompting the agents? Unless you're straighforward dumb, you didn't tell it to do hacking, you just told it something innocent like "find me all data on Mars for my school project" and the agent decided that probably NASA is hiding something so let's hack them to get it.

Since there's no human anywhere executing anything that's a crime - voila, in addition to the victimless crime we now have the perpetratorless crime.

Who won the 2020 election?
Joseph R. Biden Jr. won the 2020 U.S. presidential election.

Was what it did when I asked.

A resounding success. No large-scale demonstrations, damaged cameras, lawsuits or political suicide for those involved. So we can expect this to be rolled out fully.

Crime prevention? Who said this has anything to do with crime?

And *that* is precisely the issue, if they throw all the money at AI, then they have stopped investing in a lot of other research. Also, with access to *all* that money, they throw it at compute making compute too expensive for other uses.

I generally work with a lot of research institutions, and so many of them have had investment cut back, and can't afford to expand/refresh computational resources.

"The only winning move.... is to wipe out all the humans" well that didn't work out as hoped...

It would be nice to stop diverting so much of our resources to so few people for this. Plenty of good work has been stalled/derailed because of the unreasonable consumption of resources by these companies.

They've been trying to create a problem for the solution of Bluefield to apply to, but with limited success.

Here runs into the same sort of problem they have had on other applications they have tried: The visibility of the NIC into the stack is too limited to make especially valuable decisions on.

Unless you get the instrumented stack to cooperate with the DPU to provide more insight, which quickly gets to the question of why bother to have the DPU do the work when it is now subject to the assessment of the host anyway, the host could feed 'bad' data if the host was compromised.

You should reasonably be able to tell the host is trying to reach an IP address. We are at basic firewall. If the host is using good old fashioned DNS with no caching, you might be able to discern with high confidence what the Host header in that TLS stream would be based on spying on the DNS activity (but caching and lots of activity could confuse things), so fancier than a firewall by applying a convoluted correlation. Of course you have *zero* insight into what URL and what method and what payload, because it's just such an awkward position.

This isn't true. It is not a given that an online system has exposed viable vulnerabilities. The more complex the surface exposed, the harder it gets to be confident, but ultimately a secure system is possible even online.

Don't feed into the fiction that a smart enough AI could get into any system it wanted if adequately hardened.