How to Count AIs: Individuation and Liability for AI Agents

Abstract

The Article diagnoses the legal problem of identifying AI agents. It distinguishes thin identification, which ties every AI action to a human principal for accountability, from thick identification, which treats AI agents as persistent units with coherent goals. It proposes the Algorithmic Corporation, or A-corp, as a legal-fictional entity that can own property, contract, and litigate while being run by AIs and owned by humans.

Citation

APA: Yonathan Arbel, Simon Goldstein, Peter N. Salib. (2026). How to Count AIs: Individuation and Liability for AI Agents. https://papers.ssrn.com/sol3/papers.cfm?abstract_id=6273198

Bluebook: Yonathan Arbel, Simon Goldstein, Peter N. Salib, How to Count AIs: Individuation and Liability for AI Agents, 2026, https://papers.ssrn.com/sol3/papers.cfm?abstract_id=6273198.

Summary (English)

# How to Count AIs: Individuation and Liability for AI Agents

## TL;DR

The Article diagnoses the legal problem of identifying AI agents. It distinguishes thin identification, which ties every AI action to a human principal for accountability, from thick identification, which treats AI agents as persistent units with coherent goals. It proposes the Algorithmic Corporation, or A-corp, as a legal-fictional entity that can own property, contract, and litigate while being run by AIs and owned by humans.

## Core Contributions

* **Thin identity:** law needs a way to tie AI actions to accountable human principals.
* **Thick identity:** direct governance of AI behavior requires stable legal units for agents that copy, split, merge, and swarm.
* **A-corp proposal:** a legal-fictional entity can connect human ownership with AI-run operations.

One-page summary

# How to Count AIs: Individuation and Liability for AI Agents — one-page summary

**Paper ID:** `ssrn-6273198`
**Year:** 2026
**Author(s):** Yonathan Arbel, Simon Goldstein, Peter N. Salib
**SSRN:** https://papers.ssrn.com/sol3/papers.cfm?abstract_id=6273198

## TL;DR

The Article diagnoses the legal problem of identifying AI agents. It distinguishes thin identification, which ties every AI action to a human principal for accountability, from thick identification, which treats AI agents as persistent units with coherent goals. It proposes the Algorithmic Corporation, or A-corp, as a legal-fictional entity that can own property, contract, and litigate while being run by AIs and owned by humans.

## Keywords

AI agents; individuation; liability; algorithmic corporation; agency law; artificial intelligence governance

## Files

- Full text: `papers/ssrn-6273198/paper.txt`
- PDF: `papers/ssrn-6273198/paper.pdf`
- Summary (EN): `papers/ssrn-6273198/summary.md`

_Auto-generated study aid. For canonical content, rely on `paper.txt`/`paper.pdf`._

Study pack

# Study pack: How to Count AIs: Individuation and Liability for AI Agents (ssrn-6273198)

- SSRN: https://papers.ssrn.com/sol3/papers.cfm?abstract_id=6273198
- Full text: `papers/ssrn-6273198/paper.txt`
- Summary (EN): `papers/ssrn-6273198/summary.md`

## Elevator pitch

The Article diagnoses the legal problem of identifying AI agents. It distinguishes thin identification, which ties every AI action to a human principal for accountability, from thick identification, which treats AI agents as persistent units with coherent goals. It proposes the Algorithmic Corporation, or A-corp, as a legal-fictional entity that can own property, contract, and litigate while being run by AIs and owned by humans.

## Keywords / concepts

AI agents; individuation; liability; algorithmic corporation; agency law; artificial intelligence governance

## Suggested questions (for RAG / study)

- What is the paper’s main claim and what problem does it solve?
- What method/data does it use (if any), and what are the main results?
- What assumptions are doing the most work?
- What are the limitations or failure modes the author flags?
- How does this connect to the author’s other papers in this corpus?

_Auto-generated study aid. For canonical content, rely on `paper.txt`/`paper.pdf`._