Universal self-replicating platform AI

[synctext]

This issue aims to provide a key building block to fix our broken Internet. The goal is to create an AI bot that provides a platform service and in principle could replace the central services offered by Facebook, Uber, and Amazon.

The requirements for our platform AI bot are to provide: personal data storage, secure information processing and respecting privacy. In prior work we create a self-replicating software robot with reinforcement learning capability. Our existing bot can self-replicate by fully automatically buying servers using Bitcoin and protect privacy by buying VPN services. The current bot can operate on our decentralised exchange and has a primitive understanding of pricing and orderbooks. It requires Bitcoins to stay alive. It is meticulously designed to be suitable as a permissionless compute service for non-violent uprising. This work started in 2016 and coming years we will evolve this work future.

Technology roadmap:

• [ ] Robot ecology is expanded with 10 VPS and 10 VPN providers
• [ ] GDPR-Facebook. Refactor existing open source social network clients, rebase them on our robot platform, and use the power of GDPR data prortability to liberate Facebook-based social network data.
  • https://source.puri.sm/liberty/social/android (new fork, lack of content)
  • https://github.com/tuskyapp/Tusky
• [ ] Design a personal data storage API with durability and replication
• [ ] Permissionless computation. Integrate generic code execution capability using IPv8 modules.
• [ ] Design a robot language. Deploy this language in our robots to accelerate learning within our reinforcement learning Q-function by gossiping updates around.
• [ ] improve intelligence. When replicating the bot has a wide range of VPS options and must balance cost-effectiveness and performance. Add intelligence to record, understand, and reason with service performance demands and VPS performance of offerings.
• [ ] Replace another central platform using proprietary AI with our Universal Self-replicating Platform AI (USPA). Ironically we replace Big Tech with robots.

[sgort]

Regarding social networks, I do recommend the Mediations on Thinkspot by Tim Pastoor https://medium.com/@2W/meditations-on-thinkspot-94114fd62fa2

[sgort]

I would also love to see Tom van Engers (UvA) being involved here. Where relevant though. https://dare.uva.nl/search?identifier=21b914c8-f97d-4d54-9d7f-705623c9b85a -> A_Normative_Agentbased_Model_for_Sharing_Data_in_Secure_Trustworthy_Digital_Market.pdf

PS We're currently working together with Tom (ie Robert van Doesburg) on a compliancy by design assignment via Finance Department (Christian Schouten) using FLINT http://www.ai.rug.nl/~verheij/AI4J/papers/AI4J_paper_15_doesburg.pdf

[synctext]

Thnx for the Tim Pastoor link. His main theory: "Instead of relying on third parties to blacklist bad actors from our personal networks we should start with tools to whitelist those we already trust, and only listen to them and those they trust."

Hard problem for users to build those lists. Its the social graph.

[synctext]

Our work is becoming more relevant to Europe. Europe wants to have "technological sovereignty". "It is not too late to achieve technological sovereignty in some critical technology areas," Ursula von der Leyen

[synctext]

5221 showed us that AI-ization of everything is a remote fantasy future. Lets reopen this when we have a DAO..