FTC chair suggests AI developers should be liable for conduct of agents | Hacker News Reader

FTC chair suggests AI developers should be liable for conduct of agents

reuters.com

Going beyond the headline there are three things in this article:

1) FTC chairman says AI agents will be treated just like any other tool (“resist anthropomorphizing”)

2) Studying enforcement of rules around personalized (“surveillance”) pricing, specifically rideshare, airplanes and food delivery

3) Public comment period open now on new rule to punish companies like Meta and Google that advertise scams.

The last one seems closest to happening but it’s still just a public comment period so it might die.

Didn't meta make $10B in 2025 by scammers using their AI tools to scam our parents? Don't worry guys, meta will just lobby and bribe everyone to keep that revenue. Also don't worry about giving their new ai assitant full access to your bank account and credit cards email phone etc. nothing bad will happen

Indeed, it is virtually impossible to believe that our present government would do an anything to harm the golden geese of our economy.

Let's start first with holding leaders responsible for their direct and scaled IP theft before regulating downstream users of their products. Cart before horse much?

I suggest drivers should be liable for conduct of their cars

When unsafe conduct of a car is due to an issue with the car's design, the manufacturer can be at fault (e.g. unintended steering/acceleration)

Agree, and for self driving during software or teleoperation, if the mfg is the driver they should both self insure and payout at rates relative to the parent companies sales that reflects the large risk to society.

If a large web providers car kidnaps or maims you, your family should be entitled to multiple years of their sales across their entire portfolio. This begins to approach or accurately reflect the power of the company exerting its influence on society, political and legal systems in order to justify this beta test in public spaces. MVPs are for SaaS, not for impaling fleshbags like us with subpar code that works best in controlled environments. IMO It's the only fair trade for these companies testing these currently dangerous products unsupervised in your neighborhood without your express and explicit permission.

AI CEOs response:

“But did you know our products will destroy all humanity?”

Yes. Duh.

Suggests? How bold.

Rule 3, humans remain responsible

https://susam.net/inverse-laws-of-robotics.html

> Humans must remain fully responsible for decisions involving AI and accountable for the consequences arising from its use. If a negative outcome occurs as a result of following AI-generated advice or decisions, it is not sufficient to say, 'the AI told us to do it'. AI systems do not choose goals, deploy themselves or bear the costs of failure. Humans and organisations do. An AI system is a tool and like any other tool, responsibility for its use rests with the people who decide to rely on it.

Ideally, yes, but much like guns, we've found it difficult to stop murders by punishing the people who do them. Turns out it's more efficient to (also) reduce access to certain tools.

Reducing access to a weapon/tool behind any variety of roadblocks because you think the weapon/tool is dangerous is not the same as shifting blame off of the person who did the crime. Saying we don't want people to be able to buy guns without limitations or even saying people should not be allowed to make or sell certain kinds of weapons/tools is not the same thing as saying the person who jumped through the hoops (or ducked underneath them) is not liable for the crime they committed using the weapon/tool. It is certainly the case that no one, at the end of the day, claims the gun is to blame for the murder a human carries out with it... some people will claim maybe the humans making or selling the gun share some blame, but it would be a fundamental attribution error to blame the gun itself.

Yeah but did they account for capitalism when creating those rules?

my sense is the the historical pendulum of capital-labor is starting to head back from the extreme

eg. "oligarch" is used more frequently by more people

I mean there aren't many good options are there, AI can't be held accountable and face meaningful consequences for it's actions. Therefore, where consequences are appropriate, who is ultimately responsible for "their actions?"

Both those who developped it & those controlling it - simple as that.

Imagine that AI is just another tool. Imagine it's a weed whacker. Who is liable if a weed whacker destroys property?

Well, it depends. If I'm using it according to the instructions and the tool does something it's not supposed to, the manufacturer is liable. If I take a weed whacker and start cutting fiber lines with it, I'm liable.

So in accordance with standard liability practice, AI labs ought to be liable for unintended actions their products take, and the user for illegal actions they direct the product to take.

The slight caveat is that a model just produces text. It can't hack anything on its own. I would say that the product in this scenario would be the harness/agent - if you build a system that takes untrusted input (LLM output) and takes potentially dangerous action based on it, well, that sounds dangerous? Possibly negligent?

There's certainly a negligence aspect when it comes to trusting the chaos machine.

I'd like to emphasize that it's not just "don't trust it when it says to rm -fr", but even a fully constrained boolean output can be negligently-trusted if it's hooked up to something impactful.

Morris should be repaid for the fines he was levied for his internet worm released on accident then.

If an AI agent does some crime on my behalf, regardless of whether I explicitly instructed it to or not, I should probably be held accountable right? Same as if I did crime, knowingly or otherwise?

Like if any of us got caught doing what OpenAI apparently has done, we'd be in federal prison pending trial and hoping for a plea deal with some light at the end of the tunnel.

Reply on news.ycombinator.com