Lex Fridman avatar

Dexa/Lex Fridman

Learn more

Understanding AI Safety

Max expresses concern over the reckless approach to artificial general intelligence that relies solely on scaling neural networks without understanding their inner workings. He advocates for a more intelligible intelligence approach, emphasizing the need for research into alternative AI techniques that allow for deeper comprehension and safety assurances. The goal is to create systems that are not only powerful but also transparent and reliable, reducing the risks associated with AI deployment.