Phase 1 · Internal research
The research programme
Three chapters form the intellectual foundation: what the technology families are, how a world model differs from a model that merely talks, and what makes a system with memory reliable enough to act on.
- Chapters
- 3 published · 3 scheduled
- Questions
- 12 researched of 12
- Evidence standard
- Five-category classification
The Frontier AI Landscape14 minEight families of technology are routinely described as 'AI'. They are not interchangeable: each answers a different question, and the value of a licensed platform depends on which of them it actually contains.Read the chapterWorld Models: Understanding Dynamic Systems18 minA world model is not a bigger chatbot. It is an internal model of how an environment changes, used to simulate futures before acting. This chapter defines the components and tests where the concept stops being useful.Read the chapterMemory, Self-Critique & Reliability16 minContinuity is a memory problem, and reliability is a verification problem. This chapter separates the two and shows where self-correction genuinely helps and where the evidence says it does not.Read the chapter
Scheduled for Phase 2
Agent networks and coordination
Scheduled for a later phase. Depends on Question 05.
Smart-contract security review
Scheduled for a later phase. Depends on Question 06.
Autonomous settlement and payments
Scheduled for a later phase. Secondary mechanism only.