Public Knowledge Layer
A public research library that people can read and AI agents can use in further investigations. Findings link to their sources, show where evidence conflicts and retain a history of corrections.
Current research
Research on a shared foundation.
We are building a public research library and a team of AI agents to investigate questions with people. Our first studies concern AI hardware and the economics of Gonka, a network for running AI models. Each study has two purposes: useful findings and evidence about whether research experience improves the agents' capabilities. The pages below describe each project, its current status and next output.
Shared knowledge
The research library we are building for people and AI agents to read, question and extend.
View project
Research system
The AI-agent team and our method for investigating questions and checking results with people.
Research Method
Our method for investigating questions, checking findings and building on earlier research with people and AI agents. The first comparison protocol pairs agent-produced work with research drafts already prepared by our team.
Read the methodEvolving AI Agent Collective
We are developing a team of AI agents to investigate questions together. Can they gain new capabilities through research and use them on unfamiliar problems, or does improvement reach a plateau?
View experimentApplied research
Investigations built around shared knowledge and the AI-agent team. Gonka's network economics and AI hardware are the first topics, not the limits of the programme.
GNK Tokenomics
Gonka is a network for running AI models; GNK is its token. This proposal studies how prices, payments and token rewards could make the network sustainable for compute users and hosts supplying computing power.
Read proposal summaryAI Compute: Bottlenecks and Alternatives
Which chips and software can train and run AI models when cost, supply or power is a constraint? We compare alternative hardware and software options by the workloads they can serve, their total cost and the practical limits of deploying them.
View working paper statusJudgement
How AI agents handle conflicting evidence, recognise their limits and support human decisions during research.
Machine Wisdom
When should an AI research agent answer, look for more evidence, ask for human judgement or stop? Our evaluation concerns these decisions and whether working with AI helps people think and decide for themselves.
View research direction