Call for Participation Motivation and Topics

Recent reviews, including editorials, have contrasted the deployment of expensive equipment (e.g., responding to fires, disasters, arms races, space flights) ver- sus cheaper robot systems to reduce threats to humans. With a future that may include quantum machines and ro- bots, the implication is that robots will be used wherever teamwork is needed in hazardous conditions on earth and in space. However, while great advances have been made with large language or world models (LLMs), fewer have been made with large action models (LAMs), and virtually none for combined LAMs, where classical and quantum-like models compete for application to any form of AI-human-machine-robot teams. Once integrated into a team, the combined world models provide a reasonable means to advance the science of teams by orchestrating team work products in the field, but they have also raised concerns (viz., threats, errors). This workshop continues to explore safety, security and governance issues arising when reasoning is coupled with operating in high-stakes, dangerous or adversarial domains for cybersecurity, opera- tions, critical infrastructure, autonomy, and dual use con- texts or threats.

We welcome interdisciplinary work on the models, methods, architectures, and operational questions that arise when humans, AI systems, machines, and robots must reason, coordinate, and act together as interdependent teams. The workshop is particularly interested in the transition from language-centered AI toward systems that connect language, world models, action, tools, physical systems, and human decision-making. A contribution need not address every part of this problem. It should advance understanding of how heterogeneous human and artificial agents can be combined, orchestrated, evaluated, monitored, or governed to accomplish shared tasks in real-world settings, particularly where uncertainty, safety, reliability, or operational consequences matter.

Topics of interest include, but are not limited to:

• Language, action, and world models: Large language models, large action models, vision-language-action systems, world models, and combined architectures that connect reasoning to action. 

• Human-AI-machine-robot orchestration: Architectures and methods for forming and coordinating interdependent teams; assigning roles, tasks, and timelines; matching agent skills to task requirements; and engineering effective team structures. 

• Communication, uncertainty, and trust: How uncertainty, confidence, intent, and other information should be communicated among teammates; development and maintenance of bidirectional trust; attention management; and information sharing across human and artificial agents. 

• Decision-making and team dynamics: Consensus and majority mechanisms, distributed decision processes, coordination across agents, conflict resolution, resilience, team structure, performance, power, and other emergent properties of interdependent systems. 

• Monitoring, control, and adaptive autonomy: AI-supported monitoring of teams, transfer of control between human and machine, intervention mechanisms, operational boundaries, and methods for keeping autonomous systems within approved tasks and operating conditions. 

• Safety, security, and reliability: Hallucinations and physical errors, cybersecurity, rogue or out-of-bounds agents, robustness, predictable performance, and failure management in hazardous, high-stakes, or adversarial environments. 

• Governance and accountability: Responsibility, oversight, traceability, risk management, operational constraints, and governance of integrated human-AI-machine systems.

• Representations and evaluation of teams: Semantic drift across agents, confidence and uncertainty propagation, taxonomies of team issues, digital twins of teams, measures of team performance, and methods for evaluating whether an orchestrated system is functioning as intended.

Workshop Paper Format 

AAAI conference style (2-3 pages; please delete AAAI copyright): https://aaai.org/wp-content/uploads/2025/07/AuthorKit26-1.zip or open format. Proceedings by Elsevier, Entropy and/or Frontiers in Physics.

Important Dates 

• October 15, 2026: Submissions open

• November 22, 2026: Paper submissions to organizers 

• December 2, 2026: Notifications sent to authors

• Tuesday, December 15, 2026: AAAI-27 Early Registration Deadline (11:59 EST)

• February 22/23, 2027 (Tuesday): Workshop #39 at AAAI-27 (Palais des congrès, Montréal, Québec)