In September 2024 we released a new textbook titled “Integrating Artificial and Human Intelligence Through Agent Oriented Systems Design.” This text provides a design approach for performing conceptual design of Human-AI Agent Teams.

While AI is often discussed as a replacement to humans, much of the cognitive work we perform as humans is too complex to be performed by a single individual. Therefore, we work in teams. In a team, individuals bring diverse knowledge, skills, and abilities into the team environment. These team members work interdependently to accomplish work that is critical enough to require greater skill, ability, capacity, and redundancy than a single entity, human or AI, can reliably provide in the time available. Given the need for interdependent teamwork, this book presents a method for integrating one or more AI agents into the team to perform work and support teamwork with humans in the environment.

The book begins by introducing Human-AI Teaming, defining the necessary terminology and background. It then addresses important considerations to scope the team to be designed, define the team’s goals and responsibilities, and then discusses teaming to help the reader understand goals and responsibilities inherent in teamwork. Through these chapters, the text discusses methods to analyze the existing work environment to establish an understanding of the work that is conducted within the target environment. Following this discussion, the book discusses opportunities that AI provides for redesign and provides a high-level discussion of various AI categories and their potential use in Human-AI Teams. The book then walks through a redesign process, discussing allocation of responsibilities within teams, providing example architectural patterns for Human-AI Teams, discussing decision making and support system considerations, as well as designing systems for exceptions to improve resilience. Finally, methods are introduced to evaluate the design, and the design is translated into requirements for the system and individual AI agents.

The entire process designs team interaction among humans and future AI agents. Designing this interaction is particularly important to enhance the speed and robustness of decisions. This book uses a Model Based Systems Modeling methodology to support the design of AI systems to support teamwork. This book was developed over the past decade to support a course in Human-Agent Teaming that was requested by the 711th Human Performance Wing and built upon research that was funded by the Air Force Office of Scientific Research. This course on Human-Agent Teaming (HFEN 665) is offered by AFIT’s Department of Systems Engineering and Management as either in-residence or online using distance learning from the AFIT Graduate School. The textbook is available from the AFIT library at https://doi-org.afit.idm.oclc.org/10.1201/9781003428183 for those who have access or at https://doi.org/10.1201/9781003428183.