Follow a real strategic question through the full arbitrIQ process — step by step, with screenshots of each stage.
You frame a binary strategic question and upload supporting documents — financials, reports, market data, websites. The Director ingests this material, clarifies the question with you if needed, and generates a structured debate plan.
The plan decomposes your question into specific dimensions — each representing a distinct analytical axis that requires independent examination. For instance, a market-entry question might yield dimensions covering demand viability, operational feasibility, financial sustainability, and strategic timing.
The number of dimensions is configurable from 1 (focused triage) to 7 (comprehensive governance).
For each dimension, two independently-trained AI models are assigned opposing roles. The Advocate defends the proposition with evidence and structured reasoning. The Opposition attacks it — surfacing counter-evidence, challenging assumptions, and stress-testing projections.
The debate iterates for 2 to 10 turns (configurable). Each turn deepens: agents respond to prior arguments, narrow the space of genuine disagreement, and are forced to engage with the strongest form of the opposing position. Neither agent can agree to disagree.
Web grounding is embedded in the debate — not a separate step. The Director, Advocate, and Opposition all have access to a live web search module. A Summarizer agent processes and verifies retrieved content. This ensures arguments are grounded in current facts, not just training data.
After all debate turns are complete, the Evaluator — a different model from both debaters — reviews the full exchange for each dimension. It assesses:
Argument quality — Did each side engage with the strongest form of the opposing position?
Evidence strength — Were claims supported by data, or by assertion?
Logical coherence — Were there internal contradictions, moved goalposts, or unaddressed rebuttals?
Residual uncertainty — What was genuinely resolved, and what remains open?
The Evaluator intervenes once per dimension — its assessment is based on the complete debate record, not on partial snapshots. This ensures that late-emerging concessions and arguments are properly weighted.
The Director synthesizes all dimension evaluations and the full debate transcripts into a comprehensive executive report. This is not a summary of opinions — it is a structured account of what survived challenge, what didn't, and what remains uncertain.
The report includes integrated scoring across dimensions, a clear recommendation with explicit conditions and caveats, and a map of residual uncertainties that the decision-maker should monitor.
The full debate transcript is available alongside the report — for audit, governance review, or deeper investigation of any dimension.
Every analysis produces a complete evidentiary package — not just a recommendation.
Decision-ready synthesis with integrated scoring, recommendation, conditions, and caveats. Typically 4–8 pages.
Complete debate record across all dimensions. Every argument, rebuttal, and evidence claim preserved for audit.
Evaluator assessments per dimension — argument quality, evidence strength, and what was resolved vs. uncertain.
Explicit identification of what the analysis could not resolve — questions that require human judgment, additional data, or monitoring.
A real analysis from a consulting engagement
"Should our company expand into the Southeast Asian market within the next 24 months?"
A consulting firm submitted financials, website, and strategic context. arbitrIQ examined 4 dimensions:
1. Market Opportunity & Demand Viability
2. Operational Feasibility & Execution Risk
3. Financial Sustainability & Break-Even
4. Strategic Timing & Resource Allocation
Recommendation: Proceed only under staged entry with regulatory validation. Expansion not recommended under current leverage without capital buffer.
The Advocate built a strong case on market growth rates and competitor whitespace. The Opposition exposed that the firm's debt-to-equity ratio left no margin for the 18-month cash-burn typical of SE Asian market entry — a risk the original proposal had not quantified.
The Evaluator noted that while demand-side arguments were well-supported, the financial sustainability dimension showed the Advocate conceding key points by turn 6, suggesting the firm's current capital structure could not absorb execution delays.
Nothing hidden. Read the full output yourself.
Run your first strategic decision through structured contradiction.
Launch arbitrIQ See Pricing