AI Safety ANZ is thrilled to announce an upcoming virtual panel event on empirical and theoretical approaches to AI Alignment. The event will feature three leading researchers in the field, Lewis Hammond, Vanessa Kosoy and Jesse Hoogland who will discuss the current research landscape and the benefits and downsides of both empirical and theoretical approaches to AI Alignment. Participants will have the option of socialising with each other and other researchers in a gather town space after the event.

As artificial intelligence continues to advance rapidly, ensuring that it aligns with human values and goals is becoming an increasingly important topic. This panel will explore different approaches to achieving AI Alignment, including empirical methods that involve training and testing AI systems in realistic environments, as well as theoretical methods that involve formal models and mathematical proofs.

PANEL
══════════════════════════

Lewis Hammond is currently serving as the Acting Executive Director at the Cooperative AI Foundation and is a DPhil Affiliate at the Future of Humanity Institute, University of Oxford. He is working on the development of formal methods for multi-agent and machine learning systems, in order to help build provably safe and beneficial AI.

Vanessa Kosoy is a Research Associate at the Machine Intelligence Research Institute and her research focuses on the mathematical formalisation of general intelligence and value alignment, using tools from computational learning theory and algorithmic information theory.

Jesse Hoogland is a Research Assistant in David Krueger's lab on AI safety and science of deep learning at the University of Cambridge and is a former Machine Learning Alignment Theory Scholar at Stanford Existential Risks Initiative. His work focuses on understanding how models generalise using Singular Learning Theory and mapping out the safety insights of Deep Learning theory.

══════════════════════════

The event is open to anyone with an interest in AI and AI Alignment, regardless of technical background. Attendees will have the opportunity to learn from and engage with experts in the field, as well as ask questions and participate in discussions.
So mark your calendars for this exciting event, which will take place virtually on the 26th of April from 12:00 to 13:00 in British Summer Time (GMT+1) / from 21:00 to 22:00 in Australian Eastern Standard Time (GMT+10).