"The time for internet preaching and demagoguery has concluded. Your job, and mine, is to seize to the ground." This has also been one of my insights this past year, and it's great to see such a good recap of why that matters

“What are human values, and how do we align to them?” Very excited to release our new paper on values alignment, co-authored with Ryan Lowe and funded by @openai. 📝: meaningalignment.org/values-and-ali…

Our approach, MGE, outperforms alternatives like CCAI by @anthropic on legitimacy in a case study, and offers robustness against ideological rhetoric. 89% even agree the winning values were fair, even if their own value didn't win!

We reconcile value conflicts by asking which values participants think are wiser than others within a context. This lets us build an alignment target we call a "moral graph". It surfaces the wisest values of a large population, without relying on an ultimate moral theory.

Super interesting work from Meaning Alignment Institute. This could help achieve two hard things: 1) Help humans understand their values and at the same time 2) help align AI closer to those human values. An important step towards more human flourishing.

Super proud of the work my team has done here 🔥🔥 ✅ Defines values concretely and practically ✅ Creates a new method for surfacing wise values in a non ideological way ✅ Creates a new alignment target ✅ First step towards Wise AI

You gotta be meaning-maxxing! ✨

Excited for this! I will also be at Esmeralda for the first week and will be hosting a meaning exploration session at Spirit of Toby Shorin's pop up clinic 💫

my friends have been working on this for years and the results are actually really good!

excited about this!

I like to take on quests with my friends

Should AI be aligned with human preferences, rewards, or utility functions? Excited to finally share a preprint that Micah Carroll Matija Hal Ashton & I have worked on for almost 2 years, arguing that AI alignment has to move beyond the preference-reward-utility nexus!

combinationsmag.com is live! ✨ A publication by RadicalxChange, exploring new ideas about economics, democracy, and the relationship between technology and power. If our societies are irreversibly technological, we must steer technology itself towards the common good, rather

Introducing: Full-Stack Alignment 🥞 A research program dedicated to co-aligning AI systems *and* institutions with what people value. It's the most ambitious project I've ever undertaken. Here's what we're doing: 🧵