80,000 Hours Podcast

In this insightful discussion, Ken Goldberg, a renowned robotics professor at UC Berkeley and co-founder of innovative startups, shares his expertise on the future of robots in our lives. He outlines why training robots is more challenging than training language models and explores engineering hurdles that must be overcome for practical applications. Ken highlights sectors like homecare, agriculture, and medicine where robots could thrive. He reassures listeners about the continuing importance of human roles in workplaces amid automation's rise.

Ezra Karger, research director at the Forecasting Research Institute and economist at the Federal Reserve Bank of Chicago, discusses the complexities of forecasting existential risks like AI and nuclear conflict. He shares insights from the Existential Risk Persuasion Tournament, where predictions from experts and superforecasters revealed striking disparities in extinction probabilities. Karger emphasizes the importance of clear reference points for informed discussions and highlights the need for better forecasting methods to navigate uncertain futures involving advanced technology.

Nathan Calvin, Senior Policy Counsel at the Center for AI Safety Action Fund, discusses California's AI bill SB 1047 and its significance in shaping AI policy. He reveals how the bill addresses safety concerns amid fears of stifling innovation. The conversation highlights the contentious reactions from the tech community and the balance between regulatory requirements and fostering development. Calvin also emphasizes the broader implications of state-level regulation on national and global standards, showcasing the interconnectedness of local laws and industry practices.

Meghan Barrett, an insect neurobiologist and the founding director of the Insect Welfare Research Society, dives into the fascinating world of insect sentience and welfare. She discusses the surprising cognitive abilities of insects, the scale of potential suffering in both wild and farmed environments, and the ethical implications of farming practices. Barrett also introduces groundbreaking research on emotional states in fruit flies and advocates for a deeper understanding of insect consciousness, challenging conventional views about their capabilities and experiences.

Nick Joseph, Head of Training at Anthropic and a co-founder, discusses AI safety policies in-depth. He outlines the Responsible Scaling Policy, emphasizing the need for safeguards as AI capabilities grow. The conversation touches on the complexities of training models and the importance of external oversight. Nick addresses the financial implications of safety testing, the need for evolving safety measures, and securing AI from potential misuse. He concludes by highlighting the vital role of independent auditing and effective governance in AI development.

In this engaging conversation, Jonathan Birch, a Philosophy professor at the London School of Economics and author of 'The Edge of Sentience,' dives into the complexities of sentience across various beings, from humans with consciousness disorders to AI. He discusses the dangers of overconfidence in assessing sentience, emphasizing historical medical practices that overlooked suffering. Birch also explores ethical implications in policy-making and the evolving understanding of sentience in species like invertebrates and neural organoids, urging a careful, evidence-based approach.

Sella Nevo, director of the Meselson Center at RAND and seasoned information scientist, dives into the critical issue of securing frontier AI models. He discusses high-stakes examples of cybersecurity breaches, emphasizing how easily model weights can be targeted by rogue states and hackers. With compelling insights on human intelligence manipulation and supply chain vulnerabilities, Sella underscores the pressing need for improved defensive strategies. He also highlights his innovative machine learning work in flood forecasting, a game changer for disaster management.

Vitalik Buterin, co-founder of Ethereum, shares his insights on the intersection of technology and governance. He discusses defensive acceleration in AI and the balancing act between innovation and caution. Buterin emphasizes the importance of diverse perspectives on AI, especially in combating misinformation and navigating the ethical implications of rapid advancements. He reflects on trust in government and the need for decentralized approaches to AI, advocating for a shared understanding that can unify both optimists and skeptics in technology's future.

Sihao Huang, a 2023 Marshall Scholar and tech policy expert at RAND, dives into the complex world of US-China AI relations. He discusses how advanced AI doesn't always need massive compute power, which poses risks if misused by China. Their AI advancements are compared to the US, shedding light on China's self-sufficiency in semiconductors and AI governance. Huang emphasizes the urgent need for cooperative regulation to prevent reckless competition and potential conflict, pushing for a better understanding of AI's impact on global security.

Annie Jacobsen, a Pulitzer Prize finalist and bestselling author, dives deep into the unsettling realities of nuclear warfare. She highlights the critical six-minute decision-making window faced by the U.S. President during a nuclear threat. Jacobsen reveals the devastating humanitarian impacts of nuclear strikes and the terrifying potential of electromagnetic pulse attacks on infrastructure. Drawing from her extensive interviews with experts, she emphasizes the stark need for transparency and effective communication among nuclear powers to avert disaster.