Simon Goldstein is a research fellow at the Center for AI Safety, and an associate professor of philosophy at Australian Catholic University. His research focuses on AI safety and AI wellbeing. He is especially interested in new types of AI systems that enrich LLMs with agential scaffolding, to create AI systems that have the semblance of a folk psychology. Simon argues that some AI systems today have wellbeing, so that it could be morally wrong to frustrate their goals. He surveys the main philosophical theories of belief and desire, and argues that according to most of these theories, some AI systems have beliefs and desires. Then he surveys the main philosophical theories of wellbeing, and argues that AI systems with beliefs and desires have wellbeing. Finally, he argues that uncertainty about AI wellbeing requires us to dramatically reassess our relationship with the intelligent systems we create. MC: Bridget Loughhead Watch on Youtube: https://www.youtube.com/watch?v=21fbCIaZPlA

Monitoring and evaluation 101: How to do it practically and purposefully for your project Sophie Gulliver As EA organisations, we are always aiming to do the most good we can with the resources we have. We know that a key part of doing this is to monitor and evaluate the impact of our work, but monitoring and evaluation can appear complex, daunting, and demanding. It can be hard to know what to do, when to do it, and how to manage this within tight resource constraints. In this talk, Sophie provides a practical guide for monitoring and evaluating, drawing on research and practical experience, with the aim to give EA organisations, whatever their cause area, more confidence to monitor and evaluate their projects. Sophie Gulliver is a monitoring and evaluation freelance consultant with over seven years of experience in monitoring, evaluation, and programme management. She previously worked as an Impact and Evaluation Policy Analyst at the Centre for Global Development in their Global Health unit, and prior to that worked for UNICEF Ethiopia and CARE International. Sophie is the co-coordinator of the Effective Altruism online M&E community of practice and has consulted with organisations of all sizes in the global health and development, animal rights and community-building sectors to improve their M&E. MC: Lucas Lewit-Mendes Watch on Youtube: https://www.youtube.com/watch?v=AD9rCVzlr3E&t=2s

This talk explores the different biorisks, and compares them across different axes. It then suggests different theories of change within different areas of biorisk, and points to some of the complexities and nuances of work within the biosecurity field. Finally, it explores some common areas of biosecurity work within EA, and suggests other promising areas for exploration. Akhil is a clinical doctor who also works in global health and biosecurity. He worked as a lead researcher for Charity Entrepreneurship on global health and biosecurity research in 2022, and currently does grant making on antimicrobial resistance (for Schmidt Futures) and violence against women and girls (for The Life You Can Save). Akhil also co-founded and is the Chair of the Board for High Impact Medicine, an organisation that helps medical students and doctors do the most good with their careers. Akhil is passionate about cause prioritisation, global health and biosecurity work Watch on Youtube: https://www.youtube.com/watch?v=rsg_R4640T0

A large-scale nuclear exchange would kill billions and with geopolitical instability the risks are rising to cold war levels. Amidst this, a new arms race to gain a strategic advantage through the integration of advanced deep learning into nuclear systems is underway. Beatrice Fihn and Alice Saltini explore the heightened risks of nuclear war, the risks of AI integration in nuclear command, control and communications (NC3) and how we can prevent calamity. Beatrice Fihn is an expert on international law and civil society mobilisation, focused on developing innovative ways to strengthen global governance in the age of rapidly emerging technologies, geopolitical tension and climate change. She is the former Executive Director of the International Campaign to Abolish Nuclear Weapons (ICAN), the 2017 Nobel Peace Prize-winning campaign coalition that works to prohibit and eliminate nuclear weapons. Ms. Fihn accepted the Nobel Peace Prize and delivered the Nobel Lecture in Oslo on behalf of the campaign. As one of the worlds foremost experts on civil society mobilisation, international law, and diplomacy and multilateralism, Ms. Fihn has worked over a decade to convince governments, parliamentarians, international organisations to come together and work collectively for global governance solutions. She has spoken at the United Nations, the European Parliament, the World Economic Forum, the Munich Security Conference and SXSW, as well as lectured at numerous universities, including Harvard, Penn, Oxford and Cambridge. Alice Saltini is a Policy Fellow for the European Leadership Network, focusing most of her work at the intersection of AI and nuclear risks. With a keen interest in AI regulation at the intersection with nuclear systems, Alice is dedicated to developing innovative policy solutions in this field. Alice seeks to understand and address challenges presented by the potential implications of advanced AI models for nuclear command and control systems. Her insights have found a platform in a number of multilateral fora. Prior to joining the ELN, Alice completed stints at the Comprehensive Nuclear-Test-Ban Treaty Organization in the External Relations, Protocol and International Cooperation Section, and worked as a Research Assistant at the James Martin Center for Nonproliferation Studies. She holds a Master's degree in Russian studies and a Post Graduate Certificate (Pg Cert) in Nonproliferation Studies from the Middlebury Institute of International Studies.

Want to know why aid orgs aren't bigger fans of GiveWell and GiveDirectly type interventions? Having run the Department for International Development (DFID) ($20b/yr budget), Rory shares what the priorities of aid orgs like and what he would like them to be. Rory speaks to how we connect the pipes between EA ideas/evidence and bureaucrats. Rory Stewart was an MP (Member of Parliament) in the British Government between 2010 and 2019. He's a senior advisor for GiveDirectly and the host of the Rest Is Politics podcast. Rory's bestselling memoir "Politics On The Edge" was published in 2023, and Rory is currently the Brady–Johnson Professor of the Practice of Grand Strategy at Yale's Jackson School of Global Affairs. Watch on Youtube: https://www.youtube.com/watch?v=n9V0zF94K_U

Exploring the power of synergy in animal advocacy, this talk delves into how investigations, legal actions, legislative efforts, and corporate outreach can be intertwined for maximal impact on farmed animal welfare. Join Mahi as he uncovers optimal ways to combine these approaches. Mahi became an angry vegan and ineffective altruist in 2005. He's made some progress since then: Starting with convincing universities to opt for cage-free eggs as a student, he's now the CEO of the Albert Schweitzer Foundation. Under his leadership, every German retailer and over 100 food companies have ditched cage eggs. He's played a crucial role in ending beak searing for laying hens across Germany and pushing the European Chicken Commitment forward. His work doesn't stop at birds; Mahi's efforts have helped retailers to set higher welfare goals for their entire aquaculture supply chains — including shrimp. For his work he was honored with the Peter Singer Prize for Strategies to Reduce the Suffering of Animals. In his spare time, he uses meditation techniques to enter non-dual states of mind which he claims is helpful for his activism. Watch on Youtube: https://www.youtube.com/watch?v=dZ8NK6_twfc

In this talk, Mary shares an overview of recent work by the DeepMind Frontier Safety team. Mary discusses 1) Evaluating frontier models for dangerous capabilities, and 2) GDM's Frontier Safety Framework. Mary is a research scientist on the Safety & Alignment team at Google DeepMind, where she works on dangerous capability evaluations and responsible scaling. She was the lead author and project co-lead of DeepMind's recent paper on "Evaluating frontier models for dangerous capabilities". Watch on Youtube: https://www.youtube.com/watch?v=ZTmRT2Hg1oM

Poverty is a scaled problem and requires solutions at scale. As programs scale, however, complexities arise which may not have been considered in the original pilot evaluation. In this talk, Neela focuses on one of those scaling complexities: policy take up and program implementation. What can we systematically understand about these factors? Neela draws not only on the research but also her experience – in the private and social impact sectors to outline what we know, and what we should try and learn in future to make effective altruism even more effective. Neela is the Executive Director of the Yale Research Initiative on Innovation and Scale (Y-RISE). Previously she was the Founding Director of the Centre for Social and Behaviour Change (CBSC) at Ashoka University, India, a center funded by the Bill & Melinda Gates Foundation. She has also consulted with several non-profits including Innovations for Poverty Action, Surgo Ventures, the Busara Center for Behavioral Economics, Noora Health and Breakthrough India. Prior to her career in global health and development, Neela spent 15+ years in the private sector, leading teams in sales, brands, consumer insights and strategy in global firms Nestle, Unilever, PepsiCo and Accenture. Neela is on the board of The Life You Can Save. Her work has appeared in Harvard Business Review, Nature, Apolitical and Behavioral Scientist and she has co-authored a book "Marketplace Dignity" published by Penn Press in June 2024. She has a PhD in Marketing (Consumer Behavior) from the Wharton School, U of Pennsylvania and an MBA from IIM Calcutta, India. Watch on Youtube: https://www.youtube.com/watch?v=d6oR6gX302o

In this session, Cecil and Roberto will discuss the importance of including the Global South in global AI governance and regulation efforts, recognizing that these regions must actively participate in shaping AI protocols to manage risks effectively. Cecil Abungu is a PhD student at the University of Cambridge, and a Research Affiliate at the Centre for the Study of Existential Risk (University of Cambridge). Roberto Tinoco is a Colombian lawyer with experience in the Government Sector and Multilateral and International Organizations, in the areas of Trade, Foreign Relations and Security and Defense. Roberto is currently working at the Observatorio de Riesgos Catastróficos Globales (ORCG) as a Science Diplomat. Watch on Youtube: https://www.youtube.com/watch?v=beiDQ_BHNLQ

The talk discusses importance, tractability and neglectedness from a market shaping economics perspective, including why socially valuable innovations may be inadequately incentivized by commercial markets. Siddhartha discusses the application of pull mechanisms which reward outputs and outcomes, such as Advance Market Commitments, to pressing global challenges such as climate change and pandemic preparedness. Finally, Siddhartha discusses the analytical issues and other challenges the MSA team is grappling with including the size and design of such mechanisms. Siddhartha Haria is Senior Policy Lead at the UChicago Market Shaping Accelerator (MSA) which is led by Nobel laureate Michael Kremer, Rachel Glennerster and Christopher Snyder. The Market Shaping Accelerator aims to develop instruments that incentivize innovation to tackle pressing global challenges such as climate change and pandemic preparedness. Prior to joining MSA, Siddhartha was an government economist at the UK Department for International Development (DFID) Watch on Youtube: https://www.youtube.com/watch?v=3YizFeslQbs