Building Safe AI - MIT AI Alignment's Riya Tyagi and Gatlen Culp

Riya and Gatlen are board members of MIT AI Alignment, an organization focused on reducing risks from advanced artificial intelligence. Join us as we map the AI safety landscape, define key risks, examine generational perspectives, and explore how we can work together to build a safer future for humanity.

Hosted by Nabil Laoudji. See below for episode links, chapters, and our socials. 🙏🙌

Episode Links:

• MIT AI Alignment
• Cambridge Boston Alignment Initiative (CBAI)
• Riya's LinkedIn
• Gatlen's LinkedIn
• Gatlen's Projects
• Slaughterbots x Future of Life Institute
• Track II Diplomacy
• Tegmark AI Safety Group
• "Nexus" by Yuval Noah Harari
• RAND Institute
• BlueDot Impact
• "If Anyone Builds It, Everyone Dies" by Eliezer Yudkowsky and Nate Soares

Chapters:

• (00:00) - Preview & Intro
• (01:48) - What Is MAIA?
• (02:47) - Why AI Safety?
• (09:26) - Trends in AI Safety Interest
• (12:46) - AI Safety Techniques: MechInterp & Beyond
• (17:15) - Model Situational Awareness
• (20:58) - Hybrid, Mixture of Experts Models
• (24:40) - Decomposing a Model & Parallels with Human Brains
• (29:00) - Private Capital for Safety Research
• (32:23) - Frontier Lab Mentorship Programs
• (34:14) - Policy Perspectives & China Competition
• (36:53) - Ways In Which AI Might Threaten Us
• (39:31) - Track 2 Diplomacy & International Collaboration Examples
• (43:13) - Slaughterbots & Dangerous Capability Demos
• (46:54) - AI-Driven Unemployment
• (52:06) - Generational Attitudes Towards AI
• (57:44) - How to Get Involved - Non Technical
• (01:01:00) - Learning Resources: BlueDot Impact, etc
• (01:01:46) - Importance of Communicators & Artists
• (01:03:50) - How to Support MAIA

Connect With Us:

• Our Newsletter
• Twitter
• LinkedIn
• Bluesky
• Feedback or questions

Platforms:

• YouTube
• Apple Podcasts
• Spotify Podcasts