AI Deliberator
Guides
How to use AI DeliberatorView all 58 guides

Getting started

Evaluate and export a Council runHow to ask a better decision questionHow to choose models for an AI councilRun your first AI councilUse the Consideration MapContinue a decision with follow-up questionsEstimate and control LLM costs

Everyday decisions

Use an AI council for a career changeEvaluate a leadership role or promotionThink through a relationship decisionDecide whether to move to a new cityCompare degrees, certificates, and education optionsAnalyze a major purchaseUse live evidence mode

Work and teams

Use deliberation for a business strategy decisionSupport a hiring decision responsiblyPrioritize a product roadmapRun a decision pre-mortemMap a stakeholder disagreementAnalyze a policy proposal

Research design

Duplicate and pilot an experimentDesign an LLM deliberation experimentDefine outcomes and dependent variablesConfigure experimental conditions and independent variablesCompare deliberation with homogeneous reconsiderationBuild an independent model ensembleUse randomization, replications, and compute matching

Participant studies

Build a participant interaction studyDesign pre- and post-interaction questionnairesChoose Likert, slider, and questionnaire formatsAdd attention, manipulation, and exposure checksAnalyze and export participant-study resultsBuild a composable experiment timelineMeasure human–AI advice relianceAdd process tracing and cognitive tasksRun a human deliberation conditionBuild a human–AI hybrid deliberation chainRun a text, image, or video stimulus experimentRun a learning and tutoring experimentPlan a powered participant studyMonitor recruitment, cost, and attrition

Behavioral games

Design a behavioral economics gameRun a Dictator GameRun an Ultimatum GameRun a Trust or Investment GameRun a Prisoner’s DilemmaRun a Public Goods GameRun a Stag Hunt GameConfigure game payoffs, incentives, and feedbackManage private game links, consent, and completion

Benchmarks and results

Select benchmark questions and subsetsRun MMLU-Pro in AI DeliberatorRun PolicyBench for public-policy reasoningRun CogToM for social cognitionCreate and read 95% confidence-interval plotsCollaborate and preserve a reproducible experiment recordAnalyze sentiment and quantitative text metricsCompare two Council runs
Workspace

Participant studies

Plan a powered participant study

Use measurement packs, a transparent sample-size approximation, a pre-launch audit, and a machine-readable variable codebook.

The basic idea

Convenient defaults should reduce setup work without pretending to replace research design. Start with an editable measurement pack, identify one primary outcome, plan recruitment from an explicit effect and attrition assumption, and preserve the final variable coding with the study.

Example

Add the persuasion and correction pack to a council-versus-solo study, plan for a standardized mean difference of 0.30 with 80 percent power and 15 percent attrition, then resolve every blocking readiness item before opening recruitment.

How to do it

  1. 1

    Choose a measurement pack that matches the construct, then edit every item so it fits the actual stimulus and population.

  2. 2

    Enter the smallest effect worth detecting, target power, and expected attrition; use the resulting recruitment total as a planning cap when its assumptions fit.

  3. 3

    Resolve blocking readiness items and review recommended pre/post, attention, and treatment-exposure checks.

  4. 4

    Save the draft and download the CSV or JSON variable codebook alongside the protocol and eventual response data.

What to watch for

  • The quick power planner is a normal-approximation for equal-size independent groups, not a substitute for clustered, longitudinal, binary, or multiplicity-aware power analysis.
  • Adding a pack does not validate a scale or make its items appropriate for a new population.
  • A forecasting probability becomes objectively scorable only after the event outcome and scoring rule are prespecified.

What a useful result looks like

The study begins recruitment with an explicit outcome hierarchy, documented measurement coding, visible design warnings, and a defensible first-pass recruitment target.