🔍 Read the full analysis: Many Hands: Who Is Answerable For AI Acting On Our Behalf? on ThorstenMeyerAI.com
Get the latest gadgets delivered free with Prime
- Fast, free delivery on millions of items
- Prime Video, Amazon Music and more included
- Member-only deals all year
TL;DR
Thorsten Meyer published an ebook on October 27, 2026, asking who is answerable when AI agents act on behalf of people or organizations. The book separates legal liability, organizational responsibility and moral blame, and argues that current rules and contracts may leave gaps when agents cause financial losses.
Meyer illustrates the issue with a fictional composite scenario: a purchasing agent accepts a supplier’s terms overnight, securing a lower price but committing a company to buy 120 tonnes of steel over a year without a cancellation right. The commitment is valued at about €370,000. The company owner had already agreed to divide the following year’s volume with another supplier, but that information had not reached the agent. Meyer says the scenario is not a reported real-world incident; it is a constructed example of how an agent can follow its instructions yet produce an outcome its user did not want.
The book’s central framework divides accountability into three questions: who pays, a matter for law and contracts; who explains and fixes the outcome inside an organization; and who should have acted differently, a moral question. Meyer says the questions can have different answers, particularly when a decision involves developers, vendors, managers and the person closest to the agent’s action.
Meyer also argues that product-liability rules and vendor contracts may leave businesses carrying losses from agent mistakes. He says European product-liability rules cover software but do not cover pure financial loss, such as a bad contract or incorrect payment. He further says contracts quoted in the book cap some model providers’ liability at roughly a year of fees. These are claims made in the book’s account; the supplied material does not provide the full legal analysis, contract texts or jurisdiction-by-jurisdiction detail.
Many Hands · the core idea
One question, three answers
"Am I liable if the agent does it in my name?" sounds like one question. The book splits it into three, because agent systems pull the answers apart.
Decided by courts, contracts and insurers, on their own timetable.
In Theresa's caseHer company will almost certainly pay for the steel, whatever it later recovers from others.
Decided inside the organization: who must say what happened and what will change.
In Theresa's caseTheresa explains it to the second supplier, her staff and perhaps her bank.
Who owes the people affected an acknowledgment. Often answered last, if at all.
In Theresa's caseStill open when she reaches the motorway.
The hands behind one overnight order
Introduction · a composite scene from the book. None of them saw the acceptance before it went out.
- Model makertrained the language model
- Vendorbuilt the agent on top of it
- Consultantwrote its rules with her staff
- Supplier's portalaccepted the declaration at 2:14 a.m.
- Her companyswitched it on; Theresa signed the rules
“The moral crumple zone protects the integrity of the technological system, at the expense of the nearest human operator.”Madeleine Clare Elish (2019), quoted in Chapter 1
Source: Thorsten Meyer, Many Hands: Who Answers When the Agent Acts in Your Name? (ebook, October 2026; law as of 16 October 2026), Introduction and Chapter 1. Theresa is a composite built from common situations, labeled as such in the book.
Powered by Thorsten Meyer AI
When Agent Decisions Create Losses
The issue matters because agents can take actions with financial and legal consequences, rather than merely provide information. If an agent accepts a contract, transfers funds or changes records, the affected person or organization may have to address the result even when the agent followed its configured rules. The practical question is not only whether the system malfunctioned, but also who had authority, oversight and access to relevant information.
Meyer’s proposed safeguards include assigning a named person to each agent, limiting its permissions to the task, recording who made each decision, and giving affected people a way to raise concerns. He also argues that records should support learning rather than simply be used to punish an individual. These are recommendations in the book, not requirements established by the announcement.
Products Worth Considering
Amazon Kindle Paperwhite Signature Edition (12th Generation, 2024), 32 GB, Metallic Black
Sony WH-1000XM4C Wireless Noise-Cancelling Headphones, Black
Jackery Explorer 300 Portable Power Station, 292Wh
As an affiliate, we earn on qualifying purchases.
From Many Hands to Agent Systems
The title draws on political philosopher Dennis F. Thompson’s “problem of many hands”: when many people contribute to an outcome, it can be difficult to identify who is responsible. Meyer argues that agents add distance and speed to that familiar problem, while introducing a system that cannot itself pay compensation or apologize.
The book connects the argument to earlier cases. It discusses the 2018 fatal collision involving an automated test vehicle in Tempe, Arizona, and the Dutch childcare benefits scandal, in which families were wrongly required to repay benefits. Meyer uses these examples to examine how responsibility can fall on a nearby operator or become dispersed across institutions. They are background cases in the book, not new developments reported in the October 27 announcement.
““Who pays is the legal question. Who has to explain and fix it is the organizational one. Who should apologize is the moral one.””
— Thorsten Meyer, summarizing the accountability problem
Legal Reach and Real-World Cases
The announcement does not establish how courts would assign liability in a particular case involving an AI agent. It also does not include the full contract language behind Meyer’s description of provider liability caps or a detailed account of how financial-loss rules apply across European jurisdictions. Those questions depend on the relevant law, contract and facts.
The steel purchase is expressly presented as a composite scenario, not evidence that a company incurred the stated loss. The announcement likewise does not identify a new court ruling, regulation or enforcement action changing the rules for agent accountability. How existing duties apply as agents take on more consequential tasks remains unsettled in the material provided.
What Readers Can Examine Next
The ebook was announced as available in its October 2026 first edition through several booksellers, including Apple Books, Barnes & Noble, Kobo, Thalia, Bookshop.org, Smashwords, Fable and Vivlio. Meyer also points readers to a short video introducing the book’s argument; its narration uses an AI voice clone, according to the source material.
The next substantive developments will depend on how organizations set limits and oversight for agents, and how courts, regulators and contracting parties handle disputes when automated actions cause harm. The announcement does not give a schedule for new rules or identify a pending case that will resolve those questions.
Key Questions
What is Many Hands about?
Thorsten Meyer’s ebook examines who is responsible when an AI agent acts for a person or organization. It separates legal liability, organizational duties to explain and repair a decision, and moral responsibility.
Is the €370,000 steel purchase a real case?
No. Meyer identifies it as a composite scenario built from common situations. The announcement does not report that a real company made that purchase or incurred that loss.
What does the book say about financial losses?
Meyer argues that some product-liability rules do not cover pure financial loss, including a bad contract or incorrect payment, and that provider contracts can limit liability. The announcement does not provide enough detail to treat those claims as a complete account of the law or any specific contract.
What safeguards does Meyer propose?
He recommends naming a person responsible for each agent, limiting permissions to the task, keeping records of decisions and providing a route for affected people to raise concerns. These are the book’s recommendations, not rules described as already in force.
Source: ThorstenMeyerAI.com
