6 Thinking Hats
How different ways of thinking can help AI-assisted discussions examine a question more fully.
Read 6 Thinking Hats ↗The AI Initiative library
Ideas worth
carrying forward.
Articles, stories and a closer look at the work. Explore the thinking behind the vision, how the projects are developing, and the earlier ideas that started the journey.
Choose a question
Follow an idea through the work.
Four short reading paths connect original thinking with current project stories and architectural detail. Start with the question that interests you; the complete historical collection follows below.
What could an AI peer become?
AI Peer: A Buddy, Mentor and More · Original essay, 2024. Begin with the personal relationship: a companion that learns alongside someone.
Continue with the current vision · Programme direction. See how that early ambition develops into connected learning, memory, exploration and execution.
What helps learning carry forward?
AI in Poker · Original essay, 2024. Follow the interest in probability, decisions and an AI companion.
Continue with Modular ANN: learning in parts · Research architecture. Examine reusable neural components and how adaptation and retention can be investigated. Then explore the modular-learning research overview.
What should a collaborator remember?
Cortex: a collaborator that remembers why · Interactive project explainer. Follow a decision as its original constraints change.
Continue with the Cortex architecture · Technical overview. Explore the responsibilities behind capturing, organising, recalling and revising memory.
How can different perspectives develop an idea?
6 Thinking Hats · Original essay, 2024. Begin with complementary ways to examine a question.
Continue with Manthan: ideas with a visible history · Project story, then the exploration architecture · Technical overview. Follow how contributions, disagreement and alternatives can retain their origins. The usefulness of an idea remains a question for evaluation.
Original essays preserve their historical perspective. Current explainers distinguish implemented mechanisms, research and future directions.
Explore the full collection35 articles, project stories, technical overviews and early concepts — plus the research-paper collection.
Articles & ideas
The original writing, with its dates and historical context preserved.
AI in Poker
A personal journey into poker, probability and the idea of an AI companion for better decisions.
Read AI in Poker ↗AI Peer: A Buddy, Mentor and More
An early vision of an AI companion that learns alongside a person and adapts to their interests.
Read AI Peer: A Buddy, Mentor and More ↗Attention is all you need - AI Persona
An exploration of attention, context and the responsibilities that give an AI persona its role.
Read Attention is all you need - AI Persona ↗AI's Journey to Expertise: A Speculative Expedition
A speculative exploration of how an AI might develop expertise through learning and experience.
Read AI's Journey to Expertise: A Speculative Expedition ↗Judgment, creativity, and strategic - 6 Hats Thinking (AI)
A longer exploration of judgement, creativity and strategic thinking through different AI perspectives.
Read Judgment, creativity, and strategic - 6 Hats Thinking (AI) ↗Latent Learning with Your AI Peer: Exploring Different States of Mind
A speculative exploration of learning, attention and the different states in which people encounter ideas.
Read Latent Learning with Your AI Peer: Exploring Different States of Mind ↗Legacy: Transferring Consciousness and Accelerating Humanity's Journey
A speculative essay about continuity, personal legacy and the preservation of knowledge.
Read Legacy: Transferring Consciousness and Accelerating Humanity's Journey ↗Matiz the Math Wiz
The original Matiz concept: bringing mathematical reasoning and AI assistance into the experience of poker.
Read Matiz the Math Wiz ↗Project stories
The questions, work and contributions behind the vision.
Poker agents & Matiz
What can intelligence learn when information is incomplete and choices have consequences?
Read Poker agents & Matiz ↗Maestro
How can specialist intelligence carry an idea through coordinated work?
Read Maestro ↗Conduir
How can each piece of work make the next one more useful?
Read Conduir ↗Cortex
What should an AI collaborator remember when the work continues?
Read Cortex ↗GraFOSa
What could a change affect beyond the code directly in view?
Read GraFOSa ↗Manthan
Can we preserve how an idea develops, including the disagreements that improve it?
Read Manthan ↗AI Review Panel
How can different AI perspectives help us examine a piece of work?
Read AI Review Panel ↗AI Engineering Workforce
How do multiple AI contributors become a team whose work can be understood?
Read AI Engineering Workforce ↗Conduir Spatial
What changes when people and AI can work within the same shared space?
Read Conduir Spatial ↗Game Creation Studio
How can more ideas become experiences people can actually enter?
Read Game Creation Studio ↗Technical overviews
Architecture guides with component responsibilities, walkthroughs, examples and the limits of each design.
Conduir: architecture for work that compounds
Conduir is a programme of cooperating capabilities. Context, coordination, action, review and governance have distinct responsibilities. The architecture asks how they can form a continuing journey, so useful work and the reasons behind it do not disappear at the next handover.
Read Conduir: architecture for work that compounds ↗Cortex: memory that keeps context and purpose together
Cortex separates capturing information, preserving its context, retrieving relevant records and caring for the resulting memory. Memory retrieval provides evidence to a task; it is a different mechanism from changing the parameters of a neural model.
Read Cortex: memory that keeps context and purpose together ↗GraFOSa: understanding the relationships around a change
GraFOSa represents code structure as related entities rather than isolated files. Questions about a change can then follow those relationships. An isolated graph alternative offers a place to examine a proposal while keeping structural analysis distinct from executing and testing software.
Read GraFOSa: understanding the relationships around a change ↗Manthan: ideas with a visible history
Manthan separates generating contributions, relating them, preserving their origins and deciding which deserve further attention. Its lineage record is useful even when the quality of an idea remains unsettled. Selecting consistently useful ideas is a separate question from producing many of them.
Read Manthan: ideas with a visible history ↗AI Review Panel: making different perspectives useful
The panel separates review from adjudication. Different models examine the same subject; their findings are then challenged, consolidated and checked against that subject. The useful output is a traceable set of dispositions, not a vote that declares the work correct.
Read AI Review Panel: making different perspectives useful ↗AI Engineering Workforce: how the work is organised
This is an operating approach around engineering, not another platform successor. It separates work planning, implementation, independent review and completion decisions. The product architecture describes what is being built; the workforce approach describes how contributors work on it.
Read AI Engineering Workforce: how the work is organised ↗Conduir Spatial: working within a shared scene
Spatial work combines a shared scene, participant interaction, proposed actions and a record of change. A shared view is only part of collaboration: participants also need to know who proposed a change, whether it was accepted and what context should survive.
Read Conduir Spatial: working within a shared scene ↗Maestro: coordinating specialist intelligence
Maestro approached coordination through specialist roles, workflow structure and continuing session context. The later Maestro Platform broadened the engineering foundation; Conduir carried forward implementation and architectural work from that history.
Read Maestro: coordinating specialist intelligence ↗Modular ANN: learning in parts
Investigate whether intelligence can acquire useful capabilities through smaller, explicitly controlled changes, while preserving the ability to examine what changed.
Read Modular ANN: learning in parts ↗Poker agents and Matiz: rules, learning and experience
Make decisions under uncertainty inspectable by separating correct game mechanics, learned behaviour, evaluation and the experience a person uses.
Read Poker agents and Matiz: rules, learning and experience ↗Game Creation Studio: reusable foundations for making games
Let creators focus on the experience they want to make while reusing common behaviour for play, input, display, saves and checking.
Read Game Creation Studio: reusable foundations for making games ↗Architecture of the vision
A view of the responsibilities needed for human possibility, continuing AI peers, collective intelligence and shared worlds to reinforce one another.
Read Architecture of the vision ↗DocuMind: the original concept architecture
An explanation of the proposed document-to-reviewed-information flow, with source support and human review at its centre.
Read DocuMind: the original concept architecture ↗Quarks: the original concept architecture
An explanation of the original idea for coordinating specialised AI assistance around a person’s goal.
Read Quarks: the original concept architecture ↗Early concepts
Earlier proposals that help explain where the journey began.
DocuMind
An early proposal for using language models to explore legal documents and find relevant information.
Read DocuMind ↗Quarks
An early proposal for bringing specialised AI assistants together around a person’s goals.
Read Quarks ↗No matching pieces. Try a different term or clear the search.
Longer technical writing
White papers
Recovered technical papers and research drafts are available in the full research collection. The selection for public presentation remains open; source papers, working drafts and published articles retain their own status.