Hacker News Reader

Comment by GodelNumbering | Hacker News Reader

Not a lawyer but I can almost guarantee that the answer is yes. If it was a 'no', many malicious parties would simply start using that loophole.

View on HN

In a civil contractual dispute you can only recover actual damages. If the contract was sent, and the other party performed work on it that had a cost for them, then most likely, yes, they would be awarded damages if you refused to compensate them for any costs incurred prior to notification that the acceptance had been sent in error.

The other outcome would be clearly inequitable: forcing the counter party to eat the loss for your irresponsible use of an AI agent.

But if I paste an image of your signature onto a contract that says you give me a million pounds, that can't be legally binding for you, can it?

No, because you are a separate individual who does not have authority to sign a contract on my behalf.

Claude isn't. It's a tool, that isn't capable of signing a contract any more that Adobe Acrobat or Photoshop is.

OP used it as a tool to sign the contract. The question would be whether they did so knowingly and intentionally, if not then whether that invalidates the contract.

If it isn't then it opens up the wonderful loophole where anyone can do anything and say "the AI did it" if there are negative consequences, and "I did it" if positive. Whether the AI actually did it or not. Got caught putting out a hit on someone? "Oh I just complained to Claude about the person and said yes when it asked if I'd like it to 'handle the problem', nothing more". The user must be held accountable.

> Claude isn't. It's a tool, that isn't capable of signing a contract any more that Adobe Acrobat or Photoshop is.

Incorrect. Claude.is far more capable. E.g. it found and applied the sig without user knowledge.

> Claude.is far more capable.

Claude isn't a legal entity. It is software.

OP is entirely responsible for the software they choose to use. Nobody else. If they misused that software, a court will not be sympathetic!

That's like saying a weapon is not capable of injuring.

Capable is capable - regardless of responsibility.

It's more like saying that a weapon is not capable of murdering someone.

No one is arguing that Claude can't stick an image onto a PDF. But that is not the same thing as signing a contract.

And Adobe Acrobat is far more capable than notepad. But that doesn't change the fact that they are tools, not individuals, and thus do not have the legal authority or ability to sign contracts.

I could write dozen lines of bash that finds for PDFs, pastes an image into them and emails it to someone. That doesn't mean that bash can sign contracts.

The bot had the authority - as agent. It had the ability - as the report itself shows.

> I could write dozen lines of bash that finds for PDFs, pastes an image into them and emails it to someone. That doesn't mean that bash can sign contracts

It does mean exactly that - though bash would need to get lucky with selection and positioning.

I mean in this specific case, it did, at your behest even. These contracts would probably be even easier to enforce than Claude-signed ones!

I disagree, in the same way that I'd disagree if you said the rollerball pen on my desk signed a contract.

They can be used by someone to sign a contract, but they cannot sign a contract themselves because they're just tools.

Your pen cannot sign a contract can purely because it doesn't have the capability to find a contract in your email, find a signature on your HD, and place that sig in the right place on that contract. If it did, then yes it would be capable of signing a contract just like Claude did in this case.

You are a legal entity that can be sued. Agent is a tool that you run and are fully responsible for.

No. It’s called forgery.

Wrong. Have you read your agents TOS? You run the agent, you accept all responsibility for what it does. You are free to sue Anthropic to try and get your money back but you already indemnified them of liability, so good luck.

The GP asked what happens if THEY did it, and they are (I give them the benefit of doubt, though you can’t be sure these days) a natural person. A clanker obviously can’t commit forgery – or any crime, being a nonperson – and sure, you can’t sue OpenAnthro Corp. if their clanker does anything unintended (which makes it utter lunacy that companies just blindly trust these things, but I digress), but that’s not what was asked.

But also, you can’t ToS your way out of criminal responsibility. If OpenAnthro Corp. offered services of human agents (remember those?!) and one of their agents committed a crime while working on a client request, no ToS in the world would exonerate them.

True, that would be forgery. I think proving whether the agent truly went rogue would be 'load bearing'.

If you did not yourself intentionally sign something, in sound mind, it’s obviously not binding. But ultimately it’s up to a court to decide if you’re telling the truth that you didn’t do it.

No, application of the principal of respondeat superior would most likely be applied to an AI agent the same as a human employee. An employer is held responsible for the actions of an employee even if it is clearly contrary to their intentions.

Fair point, but I’m not sure that applies to an employee literally forging the employer’s personal signature. And equating a user–computer program relationship with an employer–employee relationship (where there’s an actual contract to that effect) doesn’t feel right anyway. Agents are still just programs and programs cannot enter contracts (like employment) given that they are not persons. This is a computer system malfunctioning.

Edit: after thinking about it more, the relationship here is obviously that of a user and a tool, not an employer and an employee. Talking about employees is anthropomorphization.

In general the user of a potentially dangerous tool is criminally and civilly liable if the tool malfunctions – unless it can be shown that the tool was used and maintained correctly and the malfunction couldn’t reasonably have been predicted or prevented. And I’d say it would be rather difficult to argue at this point that LLMs doing unintended things couldn’t have been foreseen.

So I think I’ve changed my mind about this case. The user of free-range agents does definitely deserve any civil or criminal consequences of their reckless usage.

> given that they are not persons

A company isn't a "person", but it can enter into a contract with one. And a company is essentially a tool.

that is a specific legal fiction that has both historical precedent, and can and frequently is disregarded by courts based on the circumstances (the piercing of the corporate veil).

This opens obvious loophooles. If you had deleted all logs and trajectories, courts can't trace it. Law wasn't written for or has caught up to non-human entities capable of autonomously acting

it doesn't need to. the agent operates on your behalf, you are ultimately responsible.

Reply on news.ycombinator.com