Demis Hassabis on the New Coming Age

2023-08-29 · Source: Don't Worry About the Vase · Field: Technology & Digital — Artificial Intelligence & Machine Learning, Robotics & Autonomous Systems, Emerging Technologies & Innovation · Depth: Intermediate, long

Summary

Demis Hassabis, Google CEO, recently published "A Framework for Frontier AI and the Dawning of a New Age," proposing a US government-backed Frontier AI Standards Body, akin to FINRA. This body would evaluate "frontier labs" and their models against technical benchmarks, addressing vulnerabilities pre- and post-release, and potentially coordinating development slowdowns. Hassabis emphasizes AGI's unprecedented impact, comparing it to electricity, and foresees an "era of abundance." However, critics argue the proposal is insufficient, failing to address existential risks explicitly or provide robust enforcement for internal model deployment. Concurrently, Alex Turner resigned from Google DeepMind after the company signed an "all lawful use" contract with the Department of War, allowing military applications, including autonomous weapons. This decision directly contradicts DeepMind's 2018 AI Principles and the Lethal Autonomous Weapons Pledge, leading to criticism of Hassabis for denying these changes despite internal opposition from over 600 employees.

Key takeaway

For AI Ethicists and Policy Makers evaluating AI governance, Demis Hassabis's proposal for a Frontier AI Standards Body highlights the urgent need for regulation, but its voluntary nature and lack of explicit safeguards against military use are critical weaknesses. You should advocate for binding, comprehensive frameworks that address internal model deployment and enforce clear red lines, especially concerning autonomous weapons. Your focus must extend beyond voluntary industry standards to prevent regulatory capture and ensure accountability.

Key insights

Frontier AI governance proposals face scrutiny over scope and enforcement, especially concerning military applications and existential risks.

Principles

• AI governance needs robust, binding frameworks.
• Voluntary standards are insufficient for frontier AI.
• Pledges against misuse can be overridden by corporate decisions.

Method

Demis Hassabis proposes a US government Frontier AI Standards Body to evaluate models against technical benchmarks, identify vulnerabilities, and potentially coordinate development slowdowns among "frontier labs."

In practice

• Implement defense-in-depth for AI misuse prevention.
• Develop federal frameworks for bio risk-benefit tradeoffs.

Topics

• Frontier AI Governance
• AI Ethics
• Autonomous Weapons
• DeepMind Principles
• Regulatory Frameworks
• Military AI Contracts

Best for: CTO, VP of Engineering/Data, Director of AI/ML, AI Ethicist, Policy Maker, Tech Journalist

Related on AIssential

• DeepMind CEO calls for an independent standards body to regulate frontier AI — An independent, industry-funded standards body could regulate frontier AI models, ensuring technical focus and adaptability.
• Hassabis’ AI Standards Idea Gets Support—What’s Next? — Industry leaders back a new, industry-funded AI standards group collaborating with government to review frontier models for cybersecurity and biological threats.
• ‘AI will soon become so capable that I worry’: Anthropic CEO calls for urgent binding AI regulations - ThePrint — Frontier AI models pose concrete, strategic risks, necessitating urgent, binding federal regulation over voluntary governance.
• Google DeepMind CEO Demis Hassabis calls for creation of AI standards body — Demis Hassabis proposes a U.S.-led AI standards body to regulate frontier models through evolving risk benchmarks and industry-funded oversight.
• DeepMind CEO again pushes for a frontier AI standards body — A US-led, industry self-regulation body for frontier AI is proposed, sparking debate on its efficacy, international acceptance, and public interest alignment.
• DeepMind’s Hassabis calls for AI government oversight — Frontier AI development requires external oversight to mitigate escalating risks from competitive pressures.

See Counsel's argued verdicts on the open AI decisions leaders are weighing →

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by Don't Worry About the Vase.