Abstract: With the increasing prevalence of autonomous agents—ranging from hardware-based systems like drones to software-based agents such as chatbots—there has also been a rise in instances where these agents behave in unintended, unsafe, or even dangerous ways. This has driven growing interest in AI grounding and AI alignment. Briefly, an agent exhibiting AI grounding can demonstrate a connection between its outputs and the abstract concepts it represents, while an agent exhibiting AI alignment operates in a manner consistent with user’s intent values, norms. In this talk, I present my ongoing work on addressing these two key challenges. At its core, my research bridges classical symbolic AI with large language model (LLM) based AI. Classical AI agents are robust and provide fail-safe guarantees but suffer from the long-standing knowledge engineering bottleneck. In contrast, LLM-based AI bypasses this bottleneck but is inherently unreliable. My research aims to develop neuro-symbolic agents that combine the strengths of both approaches, ensuring reliability while reducing the need for extensive manual knowledge engineering.

Bio: Dr. Muñoz-Avila is a Program Director at the U.S. National Science Foundation's (NSF) Information and Intelligent Systems (IIS) Division, where he is cluster lead for the Information Integration and Informatics (III) program and lead for the Strengthening AI Theme for the AI Research Institutes program. Prior to joining NSF, Dr. Muñoz-Avila was a (tenured, “full”) professor of Computer Science and Engineering and of Cognitive Science at Lehigh University. He was co-director of Lehigh's Institute for Data, Intelligent Systems, and Computation (I-DISC). Dr. Muñoz-Avila is recipient of a National Science Foundation CAREER and held a Lehigh Class of 1961 Professorship. He has been chair for various international scientific meetings including the Sixth International Conference on Case-Based Reasoning (ICCBR-05) and the twenty-fifth Innovative Applications of AI Conference (IAAI-13). He was funded by the Office of Naval Research (ONR), the National Science Foundation (NSF), the Defense Advanced Research Projects Agency (DARPA), the Naval Research Laboratory (NRL) and the Air Force Research Laboratory (AFRL).