Steve’s talk on “Ethical Systems and Artificial Superintelligence” will be a broad overview of Nick Bostrom’s book “Superintelligence: Paths, Dangers, Strategies” and how normative ethics relates to Superintelligence”. It will ask questions like: what might be the goals of an artificial intelligence that can significantly out-think its human creators? Which systems of ethics might be most stable as we build and teach artificial intelligences? How viable are Asimov’s 3 laws?

SPEAKER BIO:

Steve Kommrusch is a PhD student in machine learning at Colorado State University and a former executive-level engineering fellow at Advanced Micro Devices, where he worked on industry-leading processor designs such as the new Zen CPU. As a PhD student, Steve has researched generative models for 3D medical imaging, distracted driver detection and assistance, and sequence generation of Python code. Steve also co-founded the Fort Collins MIRIx symposium on safe Artificial Intelligence.