1. The goal is to help to implement a new governance/coordination system. That would allow to minimize the risks of artificial intelligence because people could create rules that would help up to navigate the transition (or pause it, if needed). 2. The main problem with implementing a new governance/coordination system is to get a lot of users. A governance/coordination system makes sense only if there are a lot of people using it, just like social networks (e.g. Facebook) only make sense if there are a lot of people using it. 3. Plan: 1. Build a decentralized platform where people can propose, review and discuss ideas regarding: 1. What system should be used to govern and coordinate artificial intelligence? 2. What should be the rules regarding artificial intelligence? 3. Meta-discussion about the platform itself. 4. Possibly, some other topics. 2. "Decentralized" means that the data (contents, and addresses/contacts of the users) is distributed among the computers of the users, instead of being stored in a central server. It's a peer-to-peer network, not client-server, and the code is open-sourced. The goal is for everyone to be able to create a copycat, because owning attention of a large group of people (and data) is a lot of power that rather nobody should have. 3. Build a large network of people by getting some small initial user base and then having people send a link to their family, coworkers, friends and partners and them doing the same... building a large network at exponential time. The website will instruct people to share a link with their coworkers and family. 4. The initial user base can be grown simply by talking to people in a shopping mall or on the street. 5. So far, I've talked to one random person in a shopping mall and she was on board with everything - agreed with what I said, engaged in the conversation, expressed willingness to share a link with their coworkers and family and gave me a phone number to send her a link once it's ready (if it's ready). I verified that the phone number is real (not fake). 6. Possibly, grow the user base before the website is built. Before the website is built, instead of having a long text explaining the idea and what to do, make a website that is like a short interactive conversation with the user simulating the conversation that I had with the person in the shopping mall. The interactive conversation will somehow stay on the website, even after the platform is built. 7. If the people don't do what we intend them to do, ask them why and improve with that knowledge. 8. The goals of the platform: 1. Be a place for constructive sharing, reviewing and discussing ideas about governance and coordination system and rules regarding artificial intelligence. 2. Make it easy for people to start a new governance/coordination system - build a network of users that they will be able to start with. Currently, in order to start a new system, you need to grow a large network. We want to solve that problem for new system creators. 3. Ensure that the contributors/participants get reasonably rewarded for their contribution. 9. There are already social networks where you can discuss ideas (e.g. X.com), but they are not optimized for those goals, they are optimized for engagement. For example, if an ordinary person shares some idea on X.com, they just won't get any replies. 10. How to reward contribution/participation: 1. In order for people to do what we want them to do (e.g. share a link, share your ideas, review someone else's ideas, build something), we likely need to reward them somehow. 2. There exists a Nash equilibrium along the lines of "People will use this platform to achieve the goal of creating new system for the benefit of people, and ensuring that people who contribute to achievement of this goal will get rewarded (including ensuring that people will be rewarded)". It's an equilibrium because if everyone follows such strategy, then everyone has interest in complying with that strategy because they will get rewarded for that. 3. In order to establish that equilibrium, people will accept an agreement (not a legal agreement) when joining that they commit to pursue that goal. The experiments from Stag and Hunt game shows that if people can communicate on equilibrium that they will proceed with, then in 80% of cases, they are able to pursue the optimal equilibrium. Accepting that agreement is a form of communication that allows people to agree on the equilibrium. 4. Someone will then create a system through which contributors will be retroactively rewarded, because the action of creating it helps to achieve the goal of ensuring that people will be rewarded, so that action will also be rewarded. 5. It might be worth to have some contribution points so that people have information what kind of contribution is needed. 11. I have an idea how the platform would work on technical side. It rather won't use blockchain or cryptocurrency because: 1. Many people have aversion to cryptocurrency. 2. The existing solutions are not scalable (I don't have a lot of confidence that this statement is true). But I would consider that yet. 12. Anticipated problems: 1. Spam - I have some idea how to deal with spam (including people who intentionally want to destroy the platform), but at this point, I don't know if I want to share the solution (maybe I need to keep it secret). 2. GDPR - people can put some personal information into content. With truly decentralized platform, it might not be completely straightforward how to give people ability to delete their personal information, because it can be on many computers. But I suppose that there is some solution to it or I misunderstand the rules because otherwise a service like Autonomi (which stores data permanently on many computers) couldn't exist. 13. The system aims to change the way people decide things to a better way, rather than to change the existing distribution of power in the world, and therefore it's beneficial for everyone. It can also stop further increase of inequality, in a way that is beneficial to everyone. It also doesn't aim to replace the existing system but to complement it. 14. If any other team has a similar goal, then join them, if there is sufficient agreement on how to achieve that goal. For example, Torchbearer Community has a similar goal, so I'd consider joining Torchbearer and executing that plan as part of Torchbearer.