I am an academic qualified in both natural and artificial intelligence. My current scientific research focus is how technology and governance impact key human cooperative behaviour. Topics include political polarisation, economic inequality, punishment and public goods investment. I also have both more basic and more applied research programmes. My PhD (MIT) is in AI, primarily on its systems engineering, which now informs my applied work on AI transparency. I am involved in policy concerning national and transnational governance of and through digital technology. I was 1 of 9 inaugural “experts” Germany nominated to the Global Partnership of AI, where I co-chaired a committee on AI governance from 2021-2023. I am also working on antitrust and market concentration, in collaboration with domain experts.
I am highly motivated by basic scientific questions concerning natural intelligence in both individuals and collectives. Early interests include impacts of the modularity and diversity of brain structures on action selection, and on the efficacy and safety of individual learning; also, the computational roles of attention and consciousness. Understanding the adaptive utility of intelligence required understanding culture (the distribution and retention of intelligence’s outputs), public goods investment, and cooperation more generally, including punishment strategies and regulation.
As mentioned, my PhD was in Systems AI: methods for promoting the ease and safety of AI development, including transparency. Good AI devops makes autonomy easier to build, understand, control, and account for. My contributions include Behavior Oriented Design (development methodology), the ABODE real-time AI IDE, POSH action selection (AKA Behavior Trees), and several systems of synthetic emotions.