AI companies should be liable for harms caused by their deployed models

AI agents are agents of the person or organization that deploys them—and should be treated by the law as such. If a company hired human writers to write its summaries, that company would be liable for inaccuracies in those summaries. If a company’s human agent signed contracts in…

Chatbots are products, not people. They’re created and released by companies that must be held accountable when their products harm people, just like a charger that starts a fire or faulty airbags that fail to deploy in a car crash.

It is highly likely that a developer and/or deployer of an AI system would be liable for harms caused by the AI acting autonomously, unless acts of the kind in question were unforeseeable.

Like any other product, safety should come first for chatbots. And just like any other product, AI developers should be held liable when they harm the public.

I don’t care how autonomous your AI is – some human has to be responsible for what it does, or we cannot have a functioning society.

Legislate a Digital Duty of Care that puts the onus on AI companies to build in safety by design and proactively address potential harm.

Alignment can improve model behavior, and liability can discipline the organization, but neither creates the same pre-action deterrent that governs a human operator.

You want to make them bear that risk. If they do, they’ll have all the incentives they need to reduce risk.

In order for our free market to function in the public interest, the companies amassing wealth and profiting from AI must be forced to take ownership of, and internalize, these costs.

The central principle of health and safety law is that those who create risks are best placed to manage and control that risk in a sensible, proportionate, and pragmatic way.

A core theme of the legal statement is the rejection of the idea that AI has created a legal vacuum and operates outside the existing law. English law has a well-established track record of developing mechanisms for addressing technological disruption and existing doctrines in co…

Accountability generally runs to the company and its people. Who is accountable for an AI agent’s actions or omissions is one of the key questions that courts and regulators will continue to grapple with. But principles of agency, tort, and contract law, a recently enacted Califo…

Public and private organisations deploying AI remain accountable for discriminatory outputs, regardless of whether the bias arises from data, model design or system use.

When it comes to dealing with your customers, the same rules apply whether using AI or human agents.

Federal laws shouldn’t become optional just because technology is new. AI is already helping make life-altering decisions for millions of Americans, but too often, it’s operating in a gray area. My Sectoral AI Governance Act gives federal agencies clearer authority to write and e…

The legislation would apply throughout the AI lifecycle, ensuring that parties developing, providing, using or operating AI systems adopt reasonable governance measures appropriate to the level of risk involved.

Responsibility must remain clearly assigned across developers, deployers, and operators. Traceability and meaningful human oversight must ensure that autonomy never becomes an accountability gap.

At the same time, technology companies and platforms cannot hide behind the complexity of their systems. So responsibility must extend across the entire AI lifecycle, from those who design foundation models to those who deploy them to those who integrate AI into products and serv…

We need public accountability frameworks and corporate liability that follow the multi-actor chains through which AI is built and deployed, with accessible remedy for affected people wherever they live.

The meeting comes as WHO/Europe’s own assessment – the most comprehensive ever conducted on AI readiness across the WHO European Region – reveals a striking gap between ambition and governance: nearly two thirds of countries are already deploying AI in diagnostics, yet only 8% ha…

AI should be understood as an instrument whose outputs remain the responsibility of human operators rather than developers.

I just don’t think the developer is in a position to know exactly how their product is being used. Just as Microsoft isn’t held liable when a money launderer uses an Excel spreadsheet, AI companies shouldn’t be blamed when a criminal uses their technology.

Putting liability on users feels most incentive-compatible. While the link between how a model is developed and how it ends up being used is often unclear, the user decides exactly how the AI is used. Liability on users creates a strong pressure to do AI in what I consider the ri…

you are solely responsible for your screening and hiring decisions [...] You agree to use Smart Screening with appropriate human review.

SB 1012 creates an untenable liability regime that will stifle AI innovation and harm consumers’ access to beneficial technology.

By holding developers liable for every downstream misuse, it could strangle open-source development, whereas negligence caps liability at “reasonable” care.

Open source is the heart of software innovation and this bill slows it down. [...] There should be zero liability at the model layer.

SB-1047 holds liable [...] the original developer of that model. It is impossible [...] to predict every possible use. SB-1047 will force developers to pull back [...].

Under this standard, a company updating safety features or fixing security vulnerabilities becomes fully liable for completely unforeseeable consequences of those updates.

States should not be permitted to penalize AI developers for a third party’s unlawful conduct involving their models.

Making AI providers liable for securities violations generally would produce inferior incentives. [...] Neither doctrine suggests universal AI provider liability for resulting harm is appropriate.

I can go to the machine, and I can have it write a horrible poem, as the Senator said, about anyone that's bad. Is that my fault, or is that the machine's fault? I think that the question becomes-- you're going to find newspapers and TV stations that are going to be using Generat…

Primary responsibility should rest with the deployer — the human or organization that put the system to use — with developers bearing secondary obligations for design safety, disclosure, and ongoing support.

The most straightforward approach to ensuring that companies develop and use AI responsibly is to place clear obligations on them, based on their role in the AI value chain, and to hold them liable when they fail to comply.

In most situations in which a harm arises, the company bringing a product to users and customers will be the one responsible for that harm. Importantly, open source developers should not be held liable for downstream misuse.

Existing legal principles, such as under tort and contract law, can be applied by our courts to assess and determine liability in appropriate cases. That being said, we share the concerns raised by the Member, and will continue to study this issue, including consulting with pract…

It’s a very thorny area. It’s uncharted territory. There’s a good argument to be made on either side.