Decentralised machine learning compute protocol. Connecting all of the world’s machine learning capable compute power that is capable of doing machine learning training work - including CPUs and GPUs - and opening its access to engineers, researchers and academics.

Key Takeaways:

• The machine learning advantage should be as accessible as electricity
• Regulation is increasing the dangerous bias and inequalities of AI
• Organic alignment of values drives autonomy 
• Protocols remove the dangerous incentives of oligopolies 
• AGI Parity within the next decade
• By 2030 conservative estimates place the compute market as 1% of GDP

2:30: Predicting Fires With Hedge Fund Algorithms

‍

5.45: Y-Combinator Accelerator

‍

7.00: Neural Networks & Computational Barriers

‍

10.30: Data Rights 

‍

16.00: Gensyn Origins & Trustless Dispute Resolution

‍

19:30: Self-Sovereignty & The Right To Build

‍

22.30: Homestead Principles & Social Contracts

‍

26.00: Cryptography As A National Threat

‍

28.00: Decentralised Incentives

‍

29.15: Regulatory Irony & Artificial Bias 

‍

31.00: Decentralised Democracy

‍

35.20: ML Compute Accessible As Electricity

‍

39.50: Paths To True AGI

‍

42.40: From Research To Execution

‍

44.30: A Thin Protocol For A Large Problem

‍

45.50: Crippling Costs Of Amazon & Azure 

‍

51.30: Future Proofing By Avoiding Shortcuts 

‍

55.00: Autonomy Needs Value Alignment

‍

1.05.00: 1% of US GDP by 20230

‍

1.13.00: A New Artificial Species

‍

1.22.30: Protocols Remove Oligopolies