🔍 Read the full analysis: My New Book Asks What Accountability Means For AI Agents on ThorstenMeyerAI.com
Get business pricing on tech for your team
- Business-only prices and quantity discounts
- Tax-exempt purchasing
- Multiple users, one account, clear invoices
TL;DR
Thorsten Meyer says his ebook, Many Hands: Who Answers When the Agent Acts in Your Name?, was released in October 2026. It examines legal, organizational and moral responsibility when agents make decisions or commitments on behalf of people and businesses. The book’s examples and arguments are the author’s; its claims about legal coverage and liability are not independently verified here.
Thorsten Meyer says his ebook Many Hands: Who Answers When the Agent Acts in Your Name? was released in October 2026, taking up the question of who is responsible when AI agents make purchases, accept terms or otherwise act for people and organizations. The book separates that question into who pays, who must explain and repair the harm, and who should have acted differently.
Meyer opens with a purchasing scenario: an agent accepts a steel supplier’s terms, securing a 7% lower price per tonne while committing a company to buy 120 tonnes over 12 months, with no cancellation right. The commitment is worth about €370,000. The owner, Theresa, had already agreed to split the coming year’s volume with another supplier, but the agent was not told. Meyer describes Theresa as a composite character, not a real person or a documented case. In the scenario, the agent follows its rules and records its actions, while the purchasing lead says the result was not what the owner wanted.
The author presents the episode as an illustration of a gap between following instructions and serving an organization’s intentions. His framework distinguishes legal liability, internal responsibility for explaining and fixing an outcome, and moral responsibility. Meyer argues that these can fall on different people, especially when an agent’s actions are shaped by its designers, deployers, organizational rules and the person closest to the final decision.
The ebook’s first edition is dated October 2026. Meyer lists Apple Books, Barnes & Noble, Kobo, Thalia, Bookshop.org, Smashwords, Fable and Vivlio as outlets. He also links to a short video about the book, which he says uses an AI voice clone for narration. The supplied announcement does not include independent reviews, sales figures or responses from the listed retailers.
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
Responsibility Across Agent Decisions
The book addresses a practical issue as agents move from generating suggestions to taking actions in someone else’s name. A system that books, buys, pays or accepts terms can create consequences even if it operates as designed. For a business, a lower price may not offset an unwanted long-term commitment; for a person affected by an automated decision, it may be hard to identify who can explain or correct it.
Meyer argues that assigning responsibility only to the person who clicked a button or monitored a system can obscure decisions made earlier, such as what authority the agent received and what information it could use. His proposed safeguards include a named person responsible for each agent, authority limited to the task, records showing who made key decisions, and a way for affected people to seek help. These are the author’s recommendations, not a description of rules adopted by a particular regulator.
The distinction between paying for a loss and explaining or preventing it matters to readers because those processes may involve different parties. A company might face a contractual dispute while its staff must handle the customer complaint and review how the system was configured. Meyer’s framework is intended to make those separate questions visible rather than treating “the human” as a complete answer.
Products Worth Considering
Samsung 27-Inch Odyssey QD-OLED G8 4K 240Hz Gaming Monitor
DJI Osmo Pocket 3 Creator Combo Vlogging Camera with Mic 2, 1 CMOS, 4K/120fps, 3-Axis Stabilization, 128GB Card
Jackery Explorer 300 Portable Power Station with 40W Solar Panel
As an affiliate, we earn on qualifying purchases.
The Cases Behind the Argument
Meyer connects the book’s title to political philosopher Dennis F. Thompson’s “problem of many hands”: when many people contribute to an outcome, responsibility can become difficult to assign. The author says AI agents add distance and speed to that problem, while making decisions that may be difficult to attribute to a single person.
The announcement points to two earlier examples. One is the 2018 fatal collision involving an automated test vehicle in Tempe, Arizona. Meyer summarizes findings attributed to the US National Transportation Safety Board: the vehicle’s system detected pedestrian Elaine Herzberg before impact but did not settle on what she was, and its design delayed emergency braking. The source says the board identified multiple contributing factors and that the vehicle’s operator pleaded guilty to endangerment. Meyer uses the case to discuss what anthropologist Madeleine Clare Elish calls a “moral crumple zone”, in which blame can concentrate on the human nearest a system’s failure.
The second example is the Dutch childcare benefits scandal, in which families were wrongly required to repay benefits. Meyer says more than 44,000 people had been recognized as affected by a government report in September 2026. He cites the former Dutch prime minister’s statement that responsibility for a system-wide failure could only be borne jointly. These examples provide context for the book’s argument; they do not establish that every automated decision creates the same legal or moral questions.
“protects the integrity of the technological system, at the expense of the nearest human operator.”
— Madeleine Clare Elish, as quoted by Meyer
Limits of the Announcement
The source is the author’s own announcement and describes the book’s arguments, examples and availability. It does not provide an independent assessment of the book or confirm how its legal analysis has been reviewed. Meyer’s statement that European product-liability rules do not cover pure financial loss is presented as his account; the announcement does not cite a legal opinion or explain how other laws, contract terms or jurisdictions may apply.
The steel purchase is explicitly a composite scenario, so it should not be read as evidence of an actual €370,000 transaction. The source also does not identify the purchasing agent, supplier or company. It remains unclear how the book handles specific jurisdictions beyond the examples named, or whether its recommendations have been adopted by any company or public body.
Readers Can Review the Ebook
Meyer lists the ebook as available through eight named retail platforms and links to a video summarizing its argument. The announcement gives no timetable for a print edition, later editions, independent review or further reporting about the book’s reception. Readers seeking to assess its legal claims will need to consult the relevant laws and jurisdiction-specific analysis; the supplied material does not resolve how liability would be allocated in a particular agent-related dispute.
Key Questions
What is Many Hands about?
It examines who is responsible when an AI agent acts on someone’s behalf, separating legal liability, organizational duties to explain and fix outcomes, and moral responsibility.
Has the steel purchase in the opening example happened?
No real transaction is identified. Meyer says Theresa is a composite character and presents the purchase as an illustrative scenario.
When was the ebook released?
The announcement dates the ebook’s first edition to October 2026 and was published on October 27, 2026.
Does the announcement establish who is legally liable for an agent’s actions?
No. The book asks how responsibility should be divided, but the announcement does not provide a ruling on a specific dispute. Legal outcomes depend on the applicable law, contracts and facts.
Where does Meyer say readers can find the ebook?
He lists Apple Books, Barnes & Noble, Kobo, Thalia, Bookshop.org, Smashwords, Fable and Vivlio.
Source: ThorstenMeyerAI.com
