Human–AI Ecosystems, Adaptive Decision-Making, and Electricity Market Design
Buch, Englisch, Format (B × H): 155 mm x 235 mm
ISBN: 978-3-032-47813-9
Verlag: Springer
As AI agents move from generating recommendations to taking consequential actions, organizations face a new governance challenge: understanding how AI behavior becomes organizational, financial, and physical reality. argues that outcomes do not arise from AI outputs alone. Instead, consequences emerge when an AI output or agent action enters a rule-governed process—such as a workflow review, market-clearing mechanism, settlement system, audit procedure, or operational environment. Drawing on a common framework that links AI agents, actions, conversion processes, outcomes, and governance controls, the book provides a new way to analyze and govern intelligent systems across domains.
Combining organizational theory, human–AI collaboration research, reinforcement learning, and electricity market design, the book examines how rules shape the behavior of adaptive agents and how those behaviors propagate through complex systems. The organizational chapters explore behavioral embedding, hidden work, calibrated reliance, meaningful human review, and governable human–AI synergy. The book also reveals why improvements in AI capabilities do not always lead to better system outcomes, and how human oversight and operating rules can change those results. The market-design chapters provide formal and computational evidence through multi-agent reinforcement-learning experiments, renewable energy bidding agents, market simulations informed by PJM operating data, and large language model (LLM) bidding agents. Together, they demonstrate why AI actions must be evaluated not only by what agents say, but by what they do and how institutional processes convert those actions into measurable consequences.
Written for researchers, advanced graduate students, and professionals in AI governance, information systems, operations research, energy systems, and public policy, the book offers both a rigorous analytical framework and a practical governance methodology. By introducing rule red-teaming and emphasizing execution over explanation, provides a timely foundation for understanding and governing the next generation of autonomous and semi-autonomous AI systems.
Zielgruppe
Research
Autoren/Hrsg.
Fachgebiete
- Wirtschaftswissenschaften Betriebswirtschaft Management Entscheidungsfindung
- Mathematik | Informatik EDV | Informatik Informatik Künstliche Intelligenz
- Wirtschaftswissenschaften Betriebswirtschaft Unternehmensforschung
- Wirtschaftswissenschaften Volkswirtschaftslehre Volkswirtschaftslehre Allgemein Wirtschaftstheorie, Wirtschaftsphilosophie
Weitere Infos & Material
.- Part I. Rules and Human–AI Ecosystem Foundations.
.- Rules, Authority, and Consequences.
.- Behavioral Embedding in Human–AI Work Ecosystems.
.- Part II. Service Workflows, Market Rules, and Conversion Evidence.
.- Governing AI-Agent Service Workflows: Routing, Review, and Hidden Work.
.- AI Agents in a Two-Settlement Market Ecosystem.
.- How Payment Rules Change Bids, Commitments, and Real-Time Risk.
.- When Real-Time Value Changes Day-Ahead Commitment.
.- Part III. Agent Architectures and Heterogeneous Ecosystems.
.- Validation, Robustness, and Reproducibility of Learned Market Strategies.
.- When Explanation and Execution Diverge: LLM Bidding Agents.
.- Part IV. Rule Red-Teaming and Deployment Design.
.- Governing Heterogeneous Human–AI Ecosystems.
.- Testing Human–AI Ecosystems Before Release: Rule Red-Teaming and Design Principles.




