Zvi Mowshowitz on sleeping on sleeper agents, and the biggest AI updates since ChatGPT - 80,000 Hours

The Nugget

Focusing on AI safety and AI policy developments, Zvi Mowshowitz advocates for a multi-faceted approach embracing both incremental policy improvements and broader societal changes to navigate AI-related existential risks effectively.

Key quotes

"We are in it to win it. We can do various things to progress our chances incrementally."
"People who do not have hope for the future will not be willing to fight for it."
"The difference between existential risk campaigners and mundane problem solvers: existential risk people are always, 'Let’s worry about both.'"

Zvi Mowshowitz emphasizes the importance of AI safety work, focusing on transparency, the necessity of safety-focused individuals working within AI labs, and policy advocates pushing for regulations that target compute and data centers to mitigate AI risks.

Recent AI policy developments, such as the UK AI Safety Summit and the Biden administration's executive order on AI, signal growing recognition of AI risks at various levels of governance, contributing to Zvi's optimism regarding the discourse on AI safety.

Despite advancements, there's concern over slow progress in useful AI product deployment and the emergence of factions opposing AI safety measures, emphasizing the need for sustained attention to both technological and societal responses to AI developments.

Through Balsa Research, Zvi aims to tackle neglected policy areas like the Jones Act, NEPA, and housing policy, advocating for actionable changes that can significantly impact societal well-being and indirectly support a healthier discourse around AI safety by fostering a hopeful future.

🌐 Understanding "Simulacra Levels" helps navigate complex social communications and align actions with reality, promoting clarity and effectiveness in advocacy and personal interactions.
🤖 Embracing incremental policy improvements alongside ambitious AI safety research can incrementally reduce existential risks while keeping broader goals in sight.
🏡 Addressing housing policy and other societal improvements is crucial not just for immediate benefits but also for cultivating a hopeful, engaged society ready to tackle future challenges, including those posed by AI.

This summary contains AI-generated information and may have important inaccuracies or omissions.