The missing layer between AI reasoning and real-world action.
reht determines whether an exact AI-mediated action is legitimate and admissible now. It is the first commercial entry into VALO’s governed path from human intent to measured enterprise action.
Reality
What is true now?
Authority
Who delegated action?
Consequence
What follows from acting or waiting?
Execution
May this exact action proceed?
Outcome
What happened and what value was realized?
Capability is not authority.
An AI system can have access to a tool without every use of that tool being legitimate. Identity and permissions are necessary, but the current purpose, evidence, state, risk and human mandate still matter.
Who is acting?
Authentication establishes identity. It does not establish legitimacy for every action.
Who may act?
Mandate, role, scope and required human approval must remain explicit.
Should it act now?
The exact action must still be justified by current evidence, context and consequence.
Not a chatbot. A governed operating system for enterprise change.
VALO carries a human purpose or idea from observed reality and proposal through consequence simulation, governance evaluation, human authority, controlled execution, receipt, measured outcome, realized value and learning.
Understand and simulate
Observe the real situation, identify capability gaps and compare acting, changing, waiting, escalating or abstaining.
Evaluate and clear
VAIG evaluates evidence, uncertainty and risk. REHT clears or refuses the exact action. Human authority remains final.
Execute, prove and measure
RACS binds execution to the cleared action, writes the receipt and connects the action to actual outcomes and realized value.
From human intent to measured value.
The operating system, the workforce and the consequence model.
VALO Factory OS
Observes work, identifies missing capabilities and helps build purpose-specific work factories from people, agents, data, methods and integrations.
VALO Governed Workforce
Role agents can observe, analyse, propose and perform bounded work without expanding their own authority. Rolepacks define scope and escalation.
Consequence Simulation Factory
Compares value, cost, delay, opportunity cost, human impact, irreversibility and uncertainty before a decision becomes real.
Every governed action leaves evidence.
{
"authority": "delegated_and_current",
"admissibility": "evaluated_under_current_state",
"decision": "ALLOW | MODIFY | DEFER | DENY | STEP_UP | HALT",
"execution_binding": "exact_action_only",
"receipt": "durable_evidence",
"outcome": "measured"
}
Start with one narrow REHT Shadow Pilot.
Connect one concrete workflow without production impact. VALO shows what it would allow, modify, defer, deny, step up or halt, then compares those recommendations with human decisions and actual outcomes.
Common questions
No. Models and agents may propose actions. VALO governs the transition from purpose and observed reality to controlled execution, receipt and measured outcome.
Model guardrails influence output. VALO evaluates whether an exact action may cross the execution boundary under the current mandate, evidence and state.
No. A screen, voice input or phone camera can provide context, but observation never creates authority.
No. The intended pattern is an external governance and execution-control layer around existing AI, agent and workflow systems.