Dan Hendrycks on Catastrophic AI Risks
Future of Life Institute Podcast
00:00
Analyzing the Risks of Malicious AI and Bioengineered Viruses
This chapter delves into the potential catastrophic and existential risks posed by the malicious use of AI, specifically focusing on the creation of bioengineered viruses with the assistance of large language models. The speakers express concern about the lack of safety measures in AI development and explore the possibility of a pandemic caused by a bio-engineered virus enabled by AI. They also discuss the risks of AI agents spreading online and highlight the need for preventive measures such as democratic institutions and legal liability laws.
Transcript
Play full episode
