Divya Siddarth, Saffron Huang, and Jasmine Wang are launching The Collective Intelligence Project via publishing our whitepaper today. You can read more about us online, but in this post we wanted to send an explicit invitation to collaborate out to the EA community, who we feel will have both have helpful synergies, skillsets, goals, and connections, whilst having productive worldview disagreements.
CIP is an incubator for new governance models for transformative technology. Our mission is to direct technological development towards the collective good. CIP will focus on the research and development of collective intelligence capabilities: decision-making technologies, processes, and institutions that expand a group’s capacity to construct and cooperate towards shared goals. We will apply these capabilities to transformative technology: technological advances with a high likelihood of significantly altering our society.
We believe that our most pressing challenges, including many cause areas prioritized by EA, can be characterized as collective intelligence challenges: pandemics, climate change, plutocracy, and catastrophic risks from technology all require better ways to set and execute on priorities.
We are especially interested in building better CI systems with, and for, transformative AI capabilities.
Where we think we have overlap with EA here: we agree that we need effective decision-making structures that require intentional design beyond just ‘incorporate people’s input’ or having ‘experts make good decisions’. We need a plurality of governance models; we’re over-reliant on just a few systems of value and risk.
We prioritize pilots over papers. Here are some of our current projects:
New models of value elicitation, or ways to develop scalable processes for surfacing and combining group beliefs, goals, values, and preferences: Currently, we are testing the use of liquid democracy for creating more values-aligned recommender systems; augmenting the emerging discussion platform Narwhal with language model capabilities, and supporting consortia-based efforts from complementary projects such as Pol.is, the AI Objectives Institute, the Consilience Project, RadicalXChange, and New Public.
Remaking technology institutions: Our first pilot will be the ‘CI Corporation’: a scalable, capped-returns model for technology development and deployment. This builds from our past work on developing data intermediary institutions, frameworks for decentralized governance and metagovernance, internet standards-setting, and pandemic prevention consortia.
We want to catalyze an ecosystem of aligned governance research and development projects.
There’s much more in the whitepaper, including the conceptual framework of the transformative technology trilemma, which we believe is a useful companion to the idea of differential technological development, which you can check out here.
Ways you can help:
(Time-sensitive, expires Mon Feb 6 EOD) Quote tweet / retweet / like our whitepaper launch tweet and sign up for future updates
Join our ongoing projects. We are looking for people to work with on research, mapping, and sensemaking. Help us create easy-to-understand, implementable and well-researched modules for each stage of the CI stack.
Propose a new pilot or experiment (we’re especially interested in working with AI organizations on our above two priorities).
Help us find funding (We were funded through a FTX regrant that we are now unsure if we can spend and are looking to raise $200K or more for our initial “preseed”. Connections and introductions to funders would be hugely helpful. With enough funding, we’re looking to create a microgrants program.)
The post seems to have duplicate passages, starting with the second instance of “CIP is an incubator for new governance models for transformative technology” (and ending with the second item in the numbered list).
Edit: it also duplicates the passage starting with “We want to catalyze an ecosystem of aligned governance research and development projects.”
Thanks Harrison! Made the corrections!
Congrats Jasmine et al :-)
Thank you Cullen!
In case you aren’t familiar with them, you might consider reaching out to people at the Canonical Debate Lab, as it seems potentially relevant to your work.
Also maybe the Cooperative AI Foundation?
Thank you! We haven’t heard of the Canonical Debate Lab, so definitely will be reaching out, and we’re already co-hosting a workshop with the Cooperative AI Foundation in May.
In my first post in this forum I argued that institutional progress is in my view the most strategic cause.
https://forum.effectivealtruism.org/posts/4viLtxnwzMawqdPum/time-consistency-for-the-ea-community-projects-that-bridge
Have you contacted the “Unite America” people? They are working for depolarizing reforms at the state level.
Good luck with the project, I will contact you in the next months.