Compliance

Liability

Legal responsibility for harm or damages.

How the laws define this

Both legal instruments agree that liability is fundamentally the state of being legally responsible or accountable for harm and damages caused by the operation of an artificial intelligence system. The most significant divergence lies in how personhood and specific triggers assign that responsibility. The bill proposing additions to the Revised Code takes an explicit stance on human accountability, declaring AI systems nonsentient, prohibiting them from obtaining legal personhood, and ensuring liability for direct or indirect harm, injury, or damage from AI operation, output, or recommendation rests strictly with human entities. By contrast, the act amending Tennessee Code Annotated, Title 1, frames liability as legal accountability for harm, damages, or legal infractions stemming from an AI system's operation, omitting specific references to AI outputs, recommendations, or the explicit restriction of liability to human entities.

Synthesised from the 2 statutory definitions below. Each one is quoted in full, with its source.

How jurisdictions differ

What is distinctive in each definition, binding law in force first.

Jurisdiction	Instrument	Status	What is distinctive
Tennessee, United States	AN ACT to amend Tennessee Code Annotated, Title 1, relative to certain conditions of personhood.	In Force	Adds context of AI by specifying legal accountability for harm, damages, or legal infractions caused by an AI system's operation, tying to human developers and operators.
Ohio, United States	Ohio HB 469 - Declare artificial intelligence systems nonsentient and prohibit legal personhood	Proposed	Widens liability to cover direct or indirect harm, injury, or damage from AI's operation, output, or recommendation, explicitly assigning it to human owners, users, and developers.

What it triggers

In Tennessee, United States, under the Act, liability ensures legal accountability remains exclusively with human developers and operators for any harm, damages, or legal infractions caused by an AI system.source
Under Ohio HB 469, if enacted, liability for any direct or indirect harm, injury, or damage caused by an AI system would rest with human owners, users, and developers.source

Definitions (2)

The state of being legally responsible for any direct or indirect harm, injury, or damage caused by the operation, output, or recommendation of an artificial intelligence system. The bill ensures this liability rests with human entities, not the AI.

To enact sections 1357.01, 1357.02, 1357.03, 1357.04, 1357.05, 1357.06, 1357.07, 1357.08, 1357.09, 1357.10, 1357.11, and 1357.12 of the Revised Code to declare artificial intelligence systems nonsentient and to prohibit them from obtaining legal personhoodDefinition 1 of 2

The state of being responsible for something, especially by law. In the context of AI, liability refers to who is legally accountable for any harm, damages, or legal infractions caused by an AI system's operation.

AN ACT to amend Tennessee Code Annotated, Title 1, relative to certain conditions of personhood.Definition 2 of 2

Not to be confused with

Responsible Party

Responsible party refers to the specific person or entity bearing liability for AI-related harm, while liability is the broader state of being legally responsible.

Owner, Developer, or Manufacturer

Owner, developer, or manufacturer identifies the human or legal entity responsible for an AI system, whereas liability is the legal responsibility they bear for harms.

Questions and answers

Who bears liability for AI harms under Tennessee law?

Under the Tennessee Act, liability for harm, damages, or legal infractions caused by an AI system rests exclusively with human developers and operators. AI systems are barred from legal personhood.

What does Ohio HB 469 say about AI liability?

Ohio HB 469 would assign liability for direct or indirect harm from AI operation, output, or recommendation to human owners, users, and developers. It declares AI nonsentient and bars legal personhood.

How do these definitions assign AI liability to humans?

Both definitions emphasize that liability for AI-caused harms stays with humans, such as developers, operators, owners, and users, by denying AI legal personhood.