CEO Of Microsoft AI: AI Is Becoming More Dangerous And Threatening! - Mustafa Suleyman

TL;DR

Mustafa Suleyman (founder of DeepMind, now CEO of Inflection AI) argues that AI is the most consequential technology ever created, with both immense upside (radical abundance, cheaper energy/food/healthcare) and existential risk (synthetic pathogens, autonomous weapons, loss of human control).

Key insights

• AI is “omniuse”: the same core technology can identify cancerous tumors or target a battlefield strike, driving both beneficial and harmful proliferation.
• The “pessimism aversion trap” prevents elites from having the tough conversations needed to respond to the coming wave; default optimism bias is a form of fear of pessimism.
• Scaling AI models 10× every year for a decade (from 2 petaflops in 2013 to 10 billion petaflops today) produced a “magical experience” of human-like conversation, but also made cutting-edge models fro