Show Notes
Audio versions of blogs and papers from BlueDot courses. By Sella Nevo et al. In this report, RAND researchers identify real-world attack methods that malicious actors could use to steal AI model weights. They propose a five-level security framework that AI companies could implement to defend against different threats, from amateur hackers to nation-state operations. Source: https://www.rand.org/pubs/research_briefs/RBA2849-1.html A podcast by BlueDot Impact.
More from BlueDot Narrated
Audio versions of blogs and papers from BlueDot courses. Original text: https://www.faruvc.org/ Author(s): Jeff Kaufman. A podcast by BlueDot Impact.
Audio versions of blogs and papers from BlueDot courses. Original text: https://blog.bluedot.org/p/what-is-ai-alignment Author(s): Adam Jones. A podcast by BlueDot Impact.
Audio versions of blogs and papers from BlueDot courses. Original text: https://www.worksinprogress.news/p/flipping-the-switch-on-far-uvc?hide_intro_popup=true Author(s): Richard Williamson. A podcast by BlueDot Impact.
Audio versions of blogs and papers from BlueDot courses. Original text: https://www.liebertpub.com/doi/10.1089/hs.2022.0126 Author(s): Nadia X. Montazeri and Jonas B. Sandbrink. A podcast by BlueDot Impact.
Audio versions of blogs and papers from BlueDot courses. Executive Summary The chance of H5N1 or one of its descendants causing a pandemic is slim, but the impact would probably be more severe than COVID-19A reasonable forecast that H5N1 causes a pandemic as bad as or worse than COVID-19 beginning in the next year is ~4%, which means the expected cost in terms of potential harms to the U.S. is at least $640 billionThe federal government can reduce that risk by: Scaling up testing &...