Events
OTTRS Speaker Series: “Ethics for Agentic Technology”
Event Start Date: Friday, September 26, 2025 - 12:00 pm
Event End Date: Friday, September 26, 2025 - 1:00 pm
Location: Virtual
UMD students, faculty, staff, alumni, and friends—join us for the OTTRS Speaker Series. (Registration Required)
Abstract:
As semi-autonomous, artificially intelligent systems (broadly, agentic technology) become more prominent in civilian and military spaces, their immense promise for good is tempered by concern over the harms they might cause. To ensure that such systems are responsibly deployed and operate with propriety, machine ethics research focuses on how to endow these systems with the perception, knowledge, and reasoning capabilities needed to interpret complex situations for prosocial interaction and intervention. Normative reasoning (evaluative, deontic) is particularly instrumental in producing prosocial behavior. In working backwards to feed reasoning capacity, we derive entailed perception and knowledge feature requirements. Should an artificial agent be unable to confidently determine the context to reason upon or an appropriate course of action to take, these same feature requirements in perception and knowledge enable it to recognize that the situation is beyond its scope, and hand off responsibility to a human partner. In this talk, we will review a series of research projects in ethics for agentic technology aimed at qualifying artificial agents to participate as prosocial actors in person-machine teams.
Bio:
Ariel Greenberg
Ariel Greenberg is a senior staff scientist and PI/PjM at the Johns Hopkins Applied Physics Laboratory. His current research focus in ethics for artificial agents leverages previous specializations in psychophysiology, behavioral M&S, and computational social and cognitive-affective science to design machines that act humanistically in defense, intelligence, and civilian contexts. Alongside his research, Ariel guest lectures for graduate courses, serves on the APL Technical Digest editorial board, and chairs the SBP-BRiMS conference steering committee.
VirtualUMD students, faculty, staff, alumni, and friends—join us for the OTTRS Speaker Series. (Registration Required)
Abstract:
As semi-autonomous, artificially intelligent systems (broadly, agentic technology) become more prominent in civilian and military spaces, their immense promise for good is tempered by concern over the harms they might cause. To ensure that such systems are responsibly deployed and operate with propriety, machine ethics research focuses on how to endow these systems with the perception, knowledge, and reasoning capabilities needed to interpret complex situations for prosocial interaction and intervention. Normative reasoning (evaluative, deontic) is particularly instrumental in producing prosocial behavior. In working backwards to feed reasoning capacity, we derive entailed perception and knowledge feature requirements. Should an artificial agent be unable to confidently determine the context to reason upon or an appropriate course of action to take, these same feature requirements in perception and knowledge enable it to recognize that the situation is beyond its scope, and hand off responsibility to a human partner. In this talk, we will review a series of research projects in ethics for agentic technology aimed at qualifying artificial agents to participate as prosocial actors in person-machine teams.
Bio:
Ariel Greenberg
Ariel Greenberg is a senior staff scientist and PI/PjM at the Johns Hopkins Applied Physics Laboratory. His current research focus in ethics for artificial agents leverages previous specializations in psychophysiology, behavioral M&S, and computational social and cognitive-affective science to design machines that act humanistically in defense, intelligence, and civilian contexts. Alongside his research, Ariel guest lectures for graduate courses, serves on the APL Technical Digest editorial board, and chairs the SBP-BRiMS conference steering committee.
Speaker(s): Ariel Greenberg