DeepMind’s Risk Chief Warns: “We Might Not Be Ready For AGI…”

The podcast explores AI governance, balancing innovation with responsible deployment, and the future of AI's impact on labor and society. Tom Lue from Google DeepMind discusses their frontier safety framework, emphasizing the need for cross-disciplinary approaches involving legal, ethical, and technical experts. He highlights the importance of collaboration between AI and humans in various sectors, advocating for policies that support both innovation and address potential risks like job displacement. Lue also touches upon the complexities of data sovereignty, advocating for globally diverse data sets to ensure AI benefits all societies, referencing Google's AI model AlphaFold as an example of technology that can address global challenges.

Outlines

Part 1: Career Path, AI Safety Basics

Part 2: Labor, Innovation, Governance

Part 3: Liability, Risks, Model Access

Part 4: Future Outlook, Values

Sign in to continue reading, translating and more.

Continue

Amplify - Dr. Ayesha Khanna

Part 1: Career Path, AI Safety Basics

AI Safety: Balancing Risk Measurement and Global Data Diversity

From Pre-Med to Global AI Policy: A Circuitous Career Path

DeepMind's Frontier Safety Framework: Addressing Severe AI Risks

Part 2: Labor, Innovation, Governance

AI's Impact on Labor: Automation, Collaboration, and Responsible Innovation

Navigating AI Safety: Industry Norms, Governance, and International Standards

Part 3: Liability, Risks, Model Access

AI Liability and Emergent Risks: Foreseeability and Situational Awareness

Open vs. Closed Models: Balancing Democratization and Risk Mitigation

Part 4: Future Outlook, Values

Bridging the Gap: Misconceptions, Values, and the Future of AI Governance

DeepMind’s Risk Chief Warns: “We Might Not Be Ready For AGI…”

Amplify - Dr. Ayesha Khanna

Part 1: Career Path, AI Safety Basics

00:00AI Safety: Balancing Risk Measurement and Global Data Diversity

AI Safety: Balancing Risk Measurement and Global Data Diversity

00:49From Pre-Med to Global AI Policy: A Circuitous Career Path

From Pre-Med to Global AI Policy: A Circuitous Career Path

07:20DeepMind's Frontier Safety Framework: Addressing Severe AI Risks

DeepMind's Frontier Safety Framework: Addressing Severe AI Risks

Part 2: Labor, Innovation, Governance

15:14AI's Impact on Labor: Automation, Collaboration, and Responsible Innovation

AI's Impact on Labor: Automation, Collaboration, and Responsible Innovation

23:08Navigating AI Safety: Industry Norms, Governance, and International Standards

Navigating AI Safety: Industry Norms, Governance, and International Standards

Part 3: Liability, Risks, Model Access

31:37AI Liability and Emergent Risks: Foreseeability and Situational Awareness

AI Liability and Emergent Risks: Foreseeability and Situational Awareness

37:13Open vs. Closed Models: Balancing Democratization and Risk Mitigation

Open vs. Closed Models: Balancing Democratization and Risk Mitigation

Part 4: Future Outlook, Values

44:11Bridging the Gap: Misconceptions, Values, and the Future of AI Governance

Bridging the Gap: Misconceptions, Values, and the Future of AI Governance