01
BOOK
The Algorithm of Dharma
Ancient Intelligence for Conscious Machines
The conceptual inquiry: what should intelligence mean when capability begins to acquire consequence?
Intelligence · Responsibility · RestraintThe Dharma Protocol · Responsible Intelligence
PROTOCOL / ORIGINThe book asked the question.The protocol turns it into architecture.
The Dharma Protocol emerged from the inquiry developed in The Algorithm of Dharma: how should increasingly capable machines remain transparent about what they know, humble about what they do not, and willing to return consequential judgment to humans?
Transparency · Humility · DialogueOrigin00 / THE SEED
It began as a question inside a book.
The Algorithm of Dharma: Ancient Intelligence for Conscious Machines began as an exploration of intelligence, responsibility, restraint and human wisdom in the age of artificial intelligence.
As the manuscript evolved, one question kept returning: ethical principles around AI are valuable, but what would it mean to place those principles inside the decision process itself?
That question became the seed of the Dharma Protocol.
The book explored the philosophy. The protocol began translating that philosophy into system behaviour.
A continuing line from inquiry to implementation.
What began as a philosophical inquiry gradually moved toward architecture, experimentation and agentic decision systems.
02
PROTOCOL
The Dharma Protocol
A governance layer for consequential AI decisions
The architectural response: translate transparency, humility and dialogue into mechanisms that can mediate action.
Pause · Trace · Escalate · Dialogue03
PLATFORM
viveka.cloud
Experimental implementation
A practical environment for testing decision mediation, uncertainty, audit traces and structured human intervention.
Decision guidance · Auditability · Human review04
AGENTIC
Dharma Decision Console
Agentic decision mediation
The next evolution: an agentic interface capable of helping systems slow down, reason visibly and involve humans before consequential execution.
Agents · Escalation · Precedent · ControlThe progression
Book→Protocol→Platform→Agentic System
Principles02 / THREE PILLARS
Responsible intelligence needs more than capability.
The Dharma Protocol is organised around three behavioural expectations that should remain visible as machine autonomy increases.
Transparency
Make the decision visible.
Expose relevant context, assumptions, reasoning traces, policy constraints, uncertainty and the path that produced a consequential recommendation or action.
Humility
Make limits visible.
A system should recognise insufficient information, model uncertainty, unfamiliar conditions and situations in which autonomous execution should stop.
Dialogue
Keep judgment interactive.
Human oversight should not be a passive approval box. Consequential decisions should support structured questioning, clarification, escalation and learning.
Architecture03 / MEDIATION LAYER
The protocol sits between capability and consequence.
The Dharma Protocol is conceived as a model-agnostic governance layer. It does not replace foundation intelligence; it mediates when and how that intelligence should become action.
Human Domain
People, institutions, values, responsibility and the real-world consequences surrounding a decision.
Dharma Protocol
Governance rules, uncertainty, traceability, escalation, dialogue and human-review logic.
Decision Mediation
The cognitive layer that determines whether to answer, recommend, pause, escalate or execute.
Foundation Intelligence
LLMs, predictive systems, agents, tools, retrieval systems and other underlying AI capabilities.
Core modules
Trace EnginePause GateUncertainty ScorerDialogue ManagerPolicy-as-CodeHuman Review OrchestratorPrecedent & Memory
The Pause04 / HUMILITY IN ACTION
Capability does not automatically create permission.
One of the central ideas of the Dharma Protocol is simple: there are moments when an intelligent system should not optimise for speed. It should slow down.
Material uncertainty
Conflicting evidence
High-consequence decisions
Policy or ethical ambiguity
Irreversible actions
Novel or unfamiliar conditions
Required human accountability
The strongest system is not the one that can always act. It is the one that knows when action requires judgment.
The Trace05 / ACCOUNTABILITY
A consequential decision should be reconstructable.
If an institution cannot understand how a decision emerged, meaningful governance becomes difficult. The protocol therefore treats traceability as part of the decision itself.
01Input
02Context
03Assumptions
04Evidence
05Uncertainty
06Policy
07Decision
08Human intervention
09Outcome
Application06 / WHERE IT MATTERS
Designed for decisions where autonomy meets accountability.
Boards & Governance
Make AI-related assumptions, risks, escalation and accountability more legible at institutional level.
Regulated Enterprises
Create auditable controls around automated and agentic decision processes.
Critical Operations
Introduce pause and human-review mechanisms before high-impact execution.
Public Systems
Support transparency, contestability and responsible human authority in consequential automated decisions.
From Protocol to Build07 / viveka.cloud
Architecture becomes meaningful when it can be tested.
viveka.cloud is the experimental environment through which the Dharma Protocol is being translated into decision mediation, traceability, uncertainty handling and human escalation.
The Dharma Decision Console extends this toward agentic systems: AI that can reason, propose and act—but also recognise when it should pause and ask.
The thread
The book asked the question. The protocol turned the question into architecture. viveka.cloud is where the architecture begins to run.