The volume collects works from two symposia, held at Stanford University, in 2015 and 2016. The central question is the relationship between autonomous systems and artificial intelligence (AI), as well as research opportunities for integrating autonomous systems and human systems (silicon agents and carbon agents).

The book contains several research domains that are traditionally discussed in topic-specific single volumes. The editors and authors attempt to place autonomy and AI into a unified framework, emphasizing integration issues. Contributions from leading scientists follow an introductory chapter.

Chapter 2, “Reexamining Computational Support for Intelligence Analysis,” discusses intelligence analysis and the strongly coupled data fusion issues. A comparative analysis of existing tool suites explores the deficiencies of existing systems. The chapter concludes with a proposal for a functional design that attempts to eliminate the major problems and may yield a more efficient and effective solution. The chapter’s interesting and useful content is its analysis of entity resolution in an environment where bit series come out of sensors and there is textual information so that both sources should be combined into one framework.

Chapter 3, “Task Allocation Using Parallelized Clustering and Auctioning Algorithms for Heterogeneous Robotic Swarms Operating on a Cloud Network,” deals with the relationship between human error and AI. One reasonable AI application approach is to correct or recover human errors. The case study utilizes autonomous vehicles and unmanned aerial vehicles. These autonomous systems are composed of robot swarms that are controlled by adequate programs to rescue shipwreck survivors. Although the case study describes an experimental prototype in the form of simulation, the demonstrated results seem promising and open up the way for further research.

In chapter 4, “Human Information Interaction, Artificial Intelligence, and Errors,” the authors define the notion of human information interaction (HII). According to the authors, AI is an active medium (proxy) that takes part in the human-machine interactions within the infosphere [1,2] and provides assistance, guidance, and automation by employing the conveyed information. The ideas in the paper are interesting for research on cognitive information systems and cognitive information science. The outlined approach may provide a starting point to build up a framework where the concept of information, the handling of human-caused errors, and the information exchange between carbon and silicon agents can be put together.

Chapter 5, “Verification Challenges for Autonomous Systems,” discusses the definition of autonomy and how it could be achieved. The authors highlight the importance of autonomous system verification, as making a system highly independent from human operators means high costs in complexity, reliability, verification, and validation of design and realization. The authors briefly describe the available tools for verification and pinpoint the actual hiatus in the services of the current tool suites. There is an overview of research programs that tackle the emerging issues. A very useful part of the chapter is the summary that contains, in a table format, the challenges from a verification viewpoint.

In chapter 6, “Conceptualizing Overtrust in Robots: Why Do People Trust a Robot That Previously Failed,” the authors present their research results: humans trust robots even in cases when robots have made mistakes. Problems related to overtrusting robots are investigated in the framework of situational analysis that was elaborated on in the authors’ previous research.

Chapter 7, “Research Considerations and Tools for Evaluating Human-Automation Interaction with Future Unmanned Systems,” analyzes the requirements for interactions between humans and automated unmanned vehicles. The requirements, established by the Department of Defense (DoD), formulate the demand to increase the autonomy within man-machine (robot) systems so that a single human operator can supervise the autonomous system. Recent research has encountered limits to accomplish the defined objectives, so the authors present a sophisticated proposal to improve the state of the art in this field.

Chapter 8, “Robots Autonomy: Some Technical Issues,” addresses society’s concerns about the threats robots may present to humankind. The authors present a thorough review, investigating recent technical solutions of robot autonomy and ethics and morality in robot design.

Chapter 9, “How Children with Autism and Machines Learn to Interact,” explores how children with autism communicate with each other and with others in their environment. The authors’ hypothesis is that autistic reasoning is simpler than the complex control and logical inference system of AI, so an investigation of team formation and human reasoning may happen in a controlled and simpler environment.

Chapter 10, “Semantic Vector Spaces for Broadening Consideration of Consequences,” considers three approaches for reasoning systems to demonstrate the problem to be solved. The authors claim that symbolic reasoning (knowledge-based systems traditionally grounded in rule inference and logic) and vector spaces as geometric and linear algebraic representations of concepts should be combined to achieve better results.

Chapter 11, “On the Road to Autonomy: Evaluating and Optimizing Hybrid Team Dynamics,” explores team dynamics in real time through exploiting neuroscience and electroencephalographic (EEG) measurements. The authors were able to give opportunities for measurements of various conditions for teams and behaviors of members. The authors analyze this new technology’s potential in the study of teams and outline future research directions.

Chapter 12 “Cyber-Security and Optimization in Smart ̺Autonomous’ Buildings,” discusses security issues related to smart buildings. It systematically explores the chances for applying technologies such as machine learning, cognitive sensing, and AI to protect, to detect threats and vulnerabilities, and perhaps to correct the damages. The authors close with their proposals for creating resilient technologies to defend smart buildings and facilities against cyberattacks. The line of thinking and collected application opportunities of AI and cognitive science-related technologies can be employed in other fields as well.

Chapter 13 “Evaluations: Autonomy and Artificial Intelligence: A Threat or Savior?” presents the authors’ research that was presented at the symposia and then evaluates the other scientific contributions in the volume. The authors’ hypothesis is that the root cause for failure of humans to forecast the results of social processes and activities is the phenomenon of interdependence, which can be formulated as mutual information in the sense of information theory.

The volume is interesting for a wide spectrum of researchers in AI, autonomous systems, information theory, and cognitive systems (info-communication, information systems). Several thought-provoking ideas provide opportunities for further analysis, brainstorming, and research.