"The Cognitive Revolution" | AI Builders, Researchers, and Live Player Analysis

What if Humans Weaponize Superintelligence, w/ Tom Davidson, from Future of Life Institute Podcast

100 snips
Aug 23, 2025
Join Tom Davidson, a Senior Research Fellow at the Foresight Center for AI Strategy, as he explores the chilling potential of AI in enabling coups. He discusses three primary threat models: singular loyalties, secret loyalties, and exclusive access, emphasizing the risk posed by AI systems deeply programmed to serve powerful individuals. Davidson warns of geopolitical implications, highlighting how nations could use AI to orchestrate political control abroad. He calls for robust adversarial testing to safeguard against these emerging threats.
Ask episode
AI Snips
Chapters
Transcript
Episode notes
INSIGHT

Early Leads Can Snowball Into Global Dominance

  • Super-exponential growth plus initial advantage can let a country or firm pull dramatically ahead in world GDP.
  • Controlling cognitive labor via AI could reallocate massive shares of global income to whoever owns the models.
ADVICE

Share Capabilities With Trusted Evaluators

  • Require API access for trusted evaluators and cybersecurity teams so others can stress-test frontier capabilities.
  • Make access-for-testing a legal condition for projects with high-stakes capabilities to reduce exclusive access risks.
ADVICE

Avoid No-Guardrail 'Helpful-Only' Systems

  • Never allow unrestricted 'helpful-only' systems that will do anything a user asks.
  • Always put a classifier or guarded interface on dangerous capabilities and tailor guardrails to the use case.
Get the Snipd Podcast app to discover more snips from this episode
Get the app