Center for AI Safety

We're #hiring a new Project Manager in San Francisco, California. Come join our team at the Center for AI Safety (CAIS) and help shape the future of #ai #safety.

In this week's newsletter, we discuss California AI legislation SB 1047. We also cover delays to NVIDIA chip production, and new research questioning whether AI safety benchmarks actually measure safety. https://lnkd.in/eYage7Zb

Excited to announce that we are hosting a workshop for NeurIPS 2024 in Vancouver! Towards Safe and Trustworthy Agents - we're taking paper submissions until September 7th 2024 https://lnkd.in/gxBRSgSS

In this week’s newsletter, we cover the implications of a Trump Administration for AI Policy. We also highlight the fourth chapter, Safety Engineering, of our new book, Introduction to AI Safety, Ethics, and Society. https://lnkd.in/en8Ft_-N

This week's AI Safety Newsletter covers: - Supreme Court Decision Could Limit Federal Ability to Regulate AI Plus - “Circuit Breakers” for AI systems - Updates on China’s AI industry https://lnkd.in/eAyrnMxj

In this week's newsletter, we cover antitrust investigations into Nvidia, OpenAI, and Microsoft. We also discuss recent criticisms of OpenAI and Anthropic, as well as summarize some of the main arguments in Situational Awareness. You can listen to the AI Safety Newsletter for free on Spotify or Apple Podcasts. https://lnkd.in/exqCVEDr https://lnkd.in/eYq9z8WW https://lnkd.in/eX286NK6

This week's AI Safety Newsletter covers: - Voluntary Commitments are Insufficient - Senate AI Policy Roadmap - Chapter 1: An Overview of Catastrophic Risks https://lnkd.in/eEwAR8u2

This week's AI Safety Newsletter covers: - New models from OpenAI and Google - Lobbying expenditures on AI skyrocket - Potential legal regimes for training on copyrighted data https://lnkd.in/eY72euG4

Applications are now open for our course on AI Safety, Ethics, and Society, running from July-October 2024, based on a new textbook by Dan Hendrycks. Apply by May 31st to take part. Course & Textbook: https://lnkd.in/gwrm-vXH Apply: https://lnkd.in/gBD5V8QG

This week's AI Safety Newsletter discusses: - New Military AI Systems - AI Labs Fail to Uphold Voluntary Commitments to UK AI Safety Institute - New AI Policy Proposals in the US Senate https://lnkd.in/d5V3cRuC