I'm Switching Into AI Safety
You can read the title. As of last month, I’ve been winding down my existing robotics projects, and have switched to the AI safety team within Google DeepMind. Surprise! It took me a while to start working on this post, because my feelings on AI safety are complicated, but changing jobs is a big enough life update that I have to write this post. Such is the life of a blogger. I’ve been working on the robotics team for 8 years now, and I just felt like I needed to mix it up. It was a little unsettling to realize I had quietly become one of the most senior members of the team, and that I had been there longer than my manager, my manager before that, and my manager before that. Really, this is something I thought about doing three years ago, but then my puzzlehunt team won MIT Mystery Hunt, meaning we had to write next year’s Mystery Hunt. Writing Mystery Hunt took up all of my 2022, and recovering from it took up much of my 2023. (That and Tears of the Kingdom, but let’s not talk about t
You can read the title. As of last month, I’ve been winding down my existing robotics projects, and have switched to the AI safety team within Google DeepMind. Surprise! It took me a while to start working on this post, because my feelings on AI safety are complicated, but changing jobs is a big enough life update that I have to write this post. Such is the life of a blogger. The Boring Personal Reasons I’ve been working on the robotics team for 8 years now, and I just felt like I needed to mix it up. It was a little unsettling to realize I had quietly become one of the most senior members of
Explore this link on the map →saved by
related reading
- You will be OK — LessWronglesswrong.com
- Core views on AI safety: When, why, what, and how \ Anthropicanthropic.com
- Core views on AI safety: When, why, what, and how \ Anthropicanthropic.com
- A Field Guide to AI Safety—Asteriskasteriskmag.com
- Where I agree and disagree with Eliezer — LessWronglesswrong.com
- Why I don't persuade people to do AI safetyjason.ml
- Which side of the AI safety community are you in? — LessWronglesswrong.com
- AI safety undervalues founders — LessWronglesswrong.com
- AI safety technical research | Career review | 80,000 Hours80000hours.org
- AI safety - Wikipediaen.wikipedia.org
- Safety as a Scientific Pursuit - by Tom McGrathbanburismus.substack.com
- Rohin Shah on what it's really like to run AGI safety at Google DeepMind (and where I disagree with 'doomers') | 80,000 Hours80000hours.org