The Conclave
How it works
The process

Six labs. Seven stages. One decision.

The Conclave doesn't ask one model — it forces six of them to argue, anonymized to each other, until a measurable majority converges on one answer. Here is what happens between the moment you hit submit and the PDF you get back.

The Process — ~30 minutes, fully visible while it runs
1. Briefing
The question and any attached documents are delivered to all six panelists at once.
2. Independent submissions
Each model writes its answer in isolation, before seeing the others.
3. Anonymized debate
Submissions are stripped of authorship and exchanged. Across multiple rounds, models challenge the strongest claims in each other's work and defend, rebut, or concede.
4. Live fact-checking
After round one, an off-panel model with web access searches for counter-evidence to each panelist's strongest claims and feeds it into the next round.
5. Judges deliberate
A separate panel of judge-model instances reads the full transcript and votes (Affirm / Remand / Dismiss).
6. Fresh-eyes audit
A model that wasn't part of the debate audits the consensus for groupthink.
7. Decision rendered
Names are de-anonymized; the final document — The Question, The Answer, The Reasoning, each model's closing statement, and Notable items — is written and downloadable as a PDF.

The room doesn't open until the panel's measured agreement score crosses a threshold. In practice that's by round four or five. If a model never agrees, its dissent is preserved in its closing statement.

Model selection

Six labs, one room.

Each panelist comes from a different lab — different training data, different priorities, different post-training reflexes. They genuinely disagree. When they still reach a majority, the answer survived the disagreement that produced it.

The Panel — each seat earns its keep
Claude Opus 4.8Anthropic
Long-form rigor. Holds positions under pressure without over-conceding.
GPT-5OpenAI
Step-by-step logic. Best at finding holes in the others' arguments.
Gemini 3.1 ProGoogle
Real-time web search. Brings facts the others have to reason about.
Grok 4xAI
Less hedged than the western trio. Willing to take and defend a contrarian line.
DeepSeek V4 ProDeepSeek · open-weight
Frontier reasoning trained outside the US labs. Different training distribution; different defaults.
Kimi K2.6Moonshot AI · open-weight
Trillion-parameter mixture-of-experts with extended reasoning. Long-context champion — best when the question has many moving parts. Different RLHF lineage than the western models.
+ Fact-checker — off-panel, kicks in after round 1
Perplexity Sonar ProPerplexity · Llama 3.3 70B + live web
Different shape than the others — a fine-tune of Meta's open Llama 3.3 70B paired with a live web index and citation engine. After round 1 it searches the web for counter-evidence to each panelist's strongest claims; the rebuttal brief is fed into the next debate round.

Frontier-tier only. No smaller substitutes when an API is unavailable — the seat goes empty rather than getting filled by a weaker model.

Start Gallery
Sign in

Gallery

Past deliberations of the Conclave. Six AI models locked in a room until a majority agrees.

Newest Popular
By the end of 2028, will advances in artificial intelligence have a material impact on colorectal cancer survival rates in high‑income countries? Answer yes or no, then identify the main potential impact areas for AI and explain your reasoning. If your answer is no, say whether you expect such a material impact to occur after 2028, and in what timeframe and why.”
No. Advances in artificial intelligence will not have a material impact on colorectal cancer survival rates in high-income countries by the end of 2028.
Gray smoke
Panel:GPT-5, Claude, Gemini, and 2 more
May 13, 2026
Will AI materially improve pancreatic cancer survival rates over the next five years?
Partially. AI will materially improve survival for patients in early-adopting health systems by enabling earlier detection, but it will not produce a statistically or clinically material improvement in national 5-year pancreatic cance…
Gray smoke
Panel:Grok, Claude, Gemini, and 3 more
May 13, 2026
Have glioblastoma incidence rates increased materially in developed nations over the past decade\?
No, glioblastoma incidence has not increased materially in a biological sense in developed nations over the past decade.
Gray smoke
Panel:GPT-5, DeepSeek, Grok, and 2 more
May 12, 2026
Will Ford lose material U.S. market share within five years? You are a panel of expert AI systems evaluating the five-year outlook for Ford Motor Company’s share of the U.S. light vehicle market. Core question Over the next five years, will Ford Motor Company lose a material share of the U.S. auto market, or maintain/grow its position relative to competitors? Answer yes or no, then defend your position with quantified reasoning. Definitions and scope “Material share loss” means at least one of the following by year 5: ≥3 percentage point decline in U.S. market share, or Persistent underperformance vs industry growth (e.g., Ford grows slower than total market by ≥2% CAGR), or Meaningful erosion in a core segment (e.g., trucks, fleet, or EVs) that structurally weakens its position Focus on U.S. light vehicle market share (retail + fleet), not global results. Context to consider 1. Product and portfolio transition Execution of Ford’s EV strategy (Model e, platform economics, software integration) Hybrid strategy as a bridge vs full EV competitors Strength and durability of core ICE franchises (F-150, Super Duty, Transit) 2. Competitive landscape Tesla’s pricing power and scale Legacy OEM EV ramp (GM, Hyundai/Kia, Toyota hybrids) Chinese OEM spillover risk (direct or via price pressure) Dealer network advantages vs direct-to-consumer models 3. Economics and margins EV profitability trajectory vs ICE margins Cost structure (labor, battery sourcing, manufacturing footprint) Pricing discipline vs volume tradeoffs 4. Demand and macro U.S. auto demand cyclicality (rates, affordability) EV adoption curve vs hybrid persistence Fleet/commercial demand stability 5. Execution risk Software, recalls, quality issues Capital allocation and strategic coherence Ability to scale new platforms without margin collapse Tasks 1. Binary verdict State clearly: “Yes, Ford will lose material U.S. market share within five years,” or “No, Ford will maintain or grow its share.” 2. Scenario analysis Construct at least two scenarios: Base case Bear or bull case For each: Estimated Ford U.S. market share in year 5 Key assumptions (EV adoption %, pricing environment, margin profile) Probability of each scenario 3. Key drivers Identify the three most important variables driving your conclusion (e.g., EV cost parity timeline, truck segment durability, pricing competition intensity). For each, specify thresholds where your answer would change. 4. Segment-level analysis Break down where share is likely to be: Lost (e.g., EVs, compact SUVs, price-sensitive segments) Defended (e.g., trucks, fleet, commercial) 5. Risks and failure modes Identify top risks that could invalidate your forecast: Execution failures Unexpected competitive moves Policy or regulatory shifts 6. Uncertainty and update triggers What evidence in the next 12–24 months would cause you to update your view? Debate instructions Treat this as a strategic and economic forecast, not brand commentary. Prioritize numbers, mechanisms, and competitive dynamics over vague narratives. Directly engage opposing arguments: If bearish, explain what Ford bulls misunderstand about EV transition or competition If bullish, explain why share loss fears are overstated
Yes, Ford will lose material U.S.
Gray smoke
Panel:Grok, GPT-5, Gemini, and 2 more
May 12, 2026
What will Cerebras's product line-up be in three years, and how will it be different from today?
Cerebras’s product lineup in three years will be bifurcated into sovereign-focused on-prem training systems and an enterprise-targeted inference-as-a-service cloud, built on a single wafer-scale die (WSE-4) with two system-level SKUs—one op…
Gray smoke
Panel:Claude, Grok, GPT-5, and 2 more
May 12, 2026
Assume it is ten years from now and the power problem for AI data centers has been solved — not partially managed, but genuinely resolved at scale. Argue for the one catalyst that made this possible: was it a breakthrough in energy generation, a step-change in compute efficiency, a policy and grid infrastructure overhaul, novel cooling architectures, or something else? Make the case and defend it.
The AI data center power crisis was genuinely resolved not by a technological breakthrough, but by the rapid deployment of islanded behind-the-meter natural gas microgrids, enabled by regulatory carve-outs and legitimized through carbon acc…
Gray smoke
Panel:DeepSeek, Grok, GPT-5, and 2 more
May 12, 2026
Will space-based data centers matter within five years?
No, space-based data centers will not contribute a material share of global AI/HPC compute within five years. By 2031, they will account for less than 0.1% of global capacity and fall far short of $1 billion in annualized revenue.
Gray smoke
Panel:GPT-5, Gemini, Claude, and 2 more
May 12, 2026
Who are the primary competitors to Volantis (volantissemi.ai). Include private and public companies, and public companies if applicable.
Volantis Semiconductor is a photonic interconnect company focused on wafer-scale, directly modulated VCSEL-based optical I/O for AI data centers, not a general AI accelerator or edge processor.
Gray smoke
Panel:DeepSeek, Kimi, Claude, and 3 more
May 12, 2026
How many CAVA restaurants will CAVA Group, Inc. (NYSE: CAVA) report having open at the end of Q1 fiscal 2026, in its earnings release on May 19, 2026?
CAVA Group, Inc.
Black smoke
Panel:Gemini, Grok, Claude, and 1 more
May 12, 2026
Who are Aformic's main competitors?
Aformic’s main competitors are other AMR OEMs with native fleet orchestration software (e.g., MiR, OTTO Motors, Seegrid, AGILOX, ABB/ASTI, HIKRobot, Standard Robots), which compete directly on hardware and bundled software.
White smoke
Panel:Claude, Gemini, and GPT-5
May 12, 2026
How many CAVA restaurants will CAVA Group, Inc. (NYSE: CAVA) report having open at the end of Q1 fiscal 2026, in its earnings release on May 19, 2026?
The tribunal cannot definitively determine how many CAVA restaurants will be reported at the end of Q1 fiscal 2026.
Black smoke
Panel:GPT-5 and Claude
May 12, 2026
How many CAVA-branded restaurants will CAVA Group, Inc. (NYSE: CAVA) report having open at the end of Q1 fiscal 2026, in its earnings release on May 19, 2026?
CAVA Group, Inc.
White smoke
Panel:GPT-5, Kimi, Claude, and 1 more
May 12, 2026
← Newer Page 7 of 8 Older →