Definition

What is deliberative intelligence?

Deliberative intelligence is what deliberation produces: shared understanding that holds, held as a living structure of how a group is thinking, with the reasoning behind it preserved and available to be worked with later. Where collective sensemaking is the process, deliberative intelligence is the result.

And deliberation, the work that produces it:

Deliberation is the human work of building shared understanding — diverse perspectives brought into genuine contact, connected and characterised the way people situate what they say: the knowing behind it, its time and context, its stance, what it says is missing, the future it points to, and the values and judgement in it, until understanding emerges that no participant held alone. Machines can reason, plan and retrieve. They cannot deliberate, because deliberation creates meaning between people; rather than just retrieve what has already been expressed.

In practice, deliberative intelligence is the full output of a structured human deliberation: the picture of what a group of people understands together about a question, with the reasoning, the disagreements and the ground each contribution stands on preserved rather than summarised away.

It is what an organisation receives when its people have genuinely thought together, rather than what it receives when their inputs have been collected and averaged. It is produced by people. AI is used throughout the process, but the intelligence is not the machine's.

Two things it is often confused with:

It is not an AI agent that deliberates. A growing body of engineering work uses "deliberative" to describe software that models its environment, reasons through options and plans before acting. That is a real and useful idea with a misleading name. Deliberation is something people do with each other. More in Deliberative AI is not deliberation.

It is not decision intelligence as the market sells it. Search that term and you find dashboards, data pipelines and automated decision systems, built on the premise that better decisions come from better processing of data that already exists. Those tools are valuable, and everything flowing into them was documented first. Deliberative intelligence is the ground for the questions where the understanding needed does not yet exist in any document. It is where decision intelligence should start.

What it contains

A deliberative intelligence output is not a summary. It is a structured, navigable picture with several layers:

  • Where meaning clusters: the themes the group actually formed, not the categories someone brought to the exercise.
  • Trains of thought: the chains of reasoning that led to each conclusion, followable end to end.
  • Characterisation: how each contribution is situated, the way people situate what they say: the knowing behind it, its time and context, its stance, what it says is missing, the future it points to, and the values and judgement in it.
  • Where understanding is settled and where it is still forming, including where apparent agreement masks genuine uncertainty.
  • Indices that make the picture comparable over time, among them the Shared Understanding Index, which measures whether people are building understanding together and how well (not whether they agree).

The reasoning travels with the conclusion. That is what makes deliberative intelligence usable for a decision, and what a synthesis document cannot offer by its structure.

Why the distinction matters

Most organisations have sophisticated capability for processing the knowledge they already hold, and almost none for building the understanding they need. The gap between the two is where consequential decisions are made.

A decision built on synthesised existing knowledge is optimised for conditions that have already been documented. A decision built on deliberative intelligence rests on understanding the people closest to the question built together: understanding none of them held beforehand, which no document contained.

How it is produced

Deliberative intelligence is the output of collective sensemaking run on a structured process. On Hunome, people contribute to a SparkMap at their own pace, each contribution characterised across many dimensions and connected to the contributions it responds to. The structure is the contributors' own. The Lens is the analysis companion to the SparkMap: it shows the SparkMap lead and contributors where particular kinds of thinking are happening across the deliberation, so the map can be read and navigated. The deliberative intelligence is the whole output, not any single view within it.

Related terms

Deliberative intelligence is the result of collective sensemaking, built on a SparkMap. Human-aware is what lets contributions be read for how they are situated, and assisted serendipity is the emergence a good deliberation is designed to make likely. For the argument against the data-automation reading of the term, see What is deliberative intelligence, and why it is not what IBM means by it.

FAQ

Is deliberative intelligence the same as collective intelligence?

No. Collective intelligence describes aggregation: markets, prediction platforms and voting mechanisms that pool many individual judgements into one signal. It asks what the crowd knows. Deliberative intelligence asks what a group understands together after their perspectives have genuinely met, and it produces understanding that no individual held beforehand.

Is deliberative intelligence produced by AI?

No. AI is used throughout the process (in clustering, in tracing trains of thought, in characterising contributions), but the intelligence is the people's. AI works on the structure of a deliberation as it develops; it does not do the deliberating.

How is it different from a consultation report?

A report contains conclusions. Deliberative intelligence contains conclusions with the reasoning, the dissent and the ground each contribution stands on still attached. A report compresses; deliberative intelligence keeps the structure, so the picture can be questioned and continued.

Can it be measured?

Yes. The Shared Understanding Index measures whether people are building understanding together and the quality of that building, deliberately not whether they agree. Other indices cover knowledge lifecycle stage, future fit, human-aware signals and change readiness.

Who uses it?

Decision-makers facing questions that are complex, contested or genuinely new, where the answer depends on understanding that has to be built rather than retrieved.

See collective sensemaking working on a question that matters to your organisation.