Risks from power-seeking AI systems - 80,000 Hours
This is a cross-post of one of our problem profiles. We’re currently posting some of our all-time best content to Substack. Read more.| 80000hours.substack.com
This is a cross-post of one of our problem profiles. We’re currently posting some of our all-time best content to Substack. Read more.| 80000hours.substack.com
As AI models rapidly advance, evaluations are racing to keep up.| TIME
Anthropic is endorsing SB 53, the California bill that governs powerful AI systems built by frontier AI developers like Anthropic.| www.anthropic.com
We’re publishing our 2024 Responsible AI Progress Report and updating our Frontier Safety Framework and AI Principles.| Google
An illustration of a sequence of events where rogue replicating agents emerge and cause harm.| metr.org
The three biggest AI companies — Anthropic, OpenAI, and DeepMind — have now all released policies designed to make their AI models less likely to go rogue or cause catastrophic damage as they approach, and eventually exceed, human capabilities. Are they good enough?| 80,000 Hours
Advanced AI systems could have massive impacts on humanity and potentially pose global catastrophic risks. There are opportunities...| 80,000 Hours
