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 5Anthropic
Long-form rigor. Holds positions under pressure without over-conceding.
GPT-5.6 SolOpenAI
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 4.5xAI
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.
Magistral MediumMistral AI
European reasoning model. Different training lineage and deliberation style from the US labs — best when you want a non-US default in the room.
+ 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
# Japanese Companies Comparative Analysis Framework Please answer for all companies (listed below) the following questions. After answering the questions, please rank each relative to the other companies on the relative strength for each question. The rankings should be 1 for the very best and the highest number for the worst on each category. Then please rank how attractive each company is versus the other companies based on the answers to the questions in totality. After this, separate all companies into one of 3 groups: 1. Above average companies 2. Average companies 3. Below average companies ## Companies - SMC - Fanuc - Mitsubishi Electric - Omron - Keyence - Daifuku - THK - Yaskawa - Kokusai - JEOL - Disco - Advantest - Lasertec - Ulvac - Screen - Tokyo Seimitsu - Tokyo Electron - Nitto Denko - Renesas - TDK - Taiyo Yuden - Kioxia - Minebea - Nitto Boseki - Hoya - Shin-Etsu - Rohm - Ibiden - Kyocera - Nidec - Socionext - Sumco - Murata - NEC - Sony - Nintendo - Hitachi - Nomura Research Institute - Fujitsu - Nikon - Resonac ## Questions ### Nature of business - Required capital intensity over time? - ROIC compared to cost of capital? - Margins versus competitors? - Direction of ROIC and margins over time? ### Competitive positioning - Nature of competitive advantage? - Number of significant competitors? - Change in number of significant competitors over time? - Potential for displacement by new technology? ### Leverage over customers - Importance of products to end customer? - Product's cost as a percentage of customer product cost or bill of materials? - Alternative products from competitors to customer? - Customer switching costs? ### Growth Potential - Addressable market size and growth rate? - Future technology and economic changes that could affect addressable market? - Potential for significant changes in share? ### Quality of management - Consistency of stated mission and goals? - Track record of financial and strategic execution versus stated goals? - Financial performance during tenure? - Degree of board independence from management?
The most attractive Japanese industrial companies are a concentrated group of 12–13 firms—primarily in semiconductor equipment, materials, and high-moat B2B automation—that possess structurally superior economics, including through-cycle RO…
White smoke
Panel:Claude, GPT-5, Magistral, and 2 more
May 18, 2026
Should I invest in this company? See attached PDF and markdown.
No, you should not invest in K1 Semiconductor at the current $15M SAFE cap.
Gray smoke
Panel:DeepSeek, Magistral, Claude, and 3 more
May 15, 2026
Develop the optimal evidence-based topical skincare regimen for maximizing skin youthfulness over the longest possible duration, for a healthy adult with no specific dermatologic conditions. Constraints Topical products only (OTC and prescription); exclude procedures, injectables, oral supplements/medications, lasers, and devices. Evidence hierarchy: prioritize RCTs, systematic reviews, and meta-analyses in peer-reviewed dermatology literature. Cite specifically. Mechanistic/in-vitro evidence may support but not substitute. Address both prevention (20s–30s starting point) and intervention (40s+). Target endpoints to address: photoaging, fine lines/wrinkles, dyspigmentation, elasticity/firmness, barrier function, texture. Deliverables AM routine — sequenced step-by-step, with specific actives, concentrations, vehicles, and frequency. PM routine — same specificity. Ranked tier list of active ingredients by strength of evidence (tretinoin, sunscreen, vitamin C, niacinamide, peptides, AHAs/BHAs, azelaic acid, growth factors, bakuchiol, etc.) with a grade of recommendation per ingredient. Ingredient interaction map — what to avoid layering and why (e.g., retinoid + AHA irritation, ascorbic acid pH conflicts, niacinamide + vitamin C myth check). Explicit callouts of trendy/marketed ingredients with weak or placebo-tier evidence. Realistic ceiling — what topicals demonstrably can and cannot achieve versus procedural alternatives. Format: structured, citation-backed, no hedging filler. Where consensus marketing claims diverge from the evidence, say so plainly.
The optimal evidence-based topical skincare regimen for maximizing skin youthfulness over the longest duration centers on daily broad-spectrum SPF 50+ and nightly prescription tretinoin (0.025–0.05%) as non-negotiable core components.
Gray smoke
Panel:Grok, GPT-5, Gemini, and 2 more
May 14, 2026
Identify the principal scientific discovery that significantly enhanced the efficacy of psoriasis therapy in the last five years. Include evidence supporting its impact and any key mechanisms involved. Don't be overly scientific.
The principal scientific advances in psoriasis therapy over the last five years are dual IL-17A/F blockade and IL-23p19-selective inhibition, which together represent a co-leadership model: one redefining the ceiling of skin clearance, the…
White smoke
Panel:Claude, Grok, Magistral, and 3 more
May 14, 2026
After its IPO today with a closing market cap of $67B, will Cerebras be worth materially more, less, or about the same in two years? Please include a probability estimate, key factors influencing the valuation, and what developments could significantly alter this outlook.
Cerebras will be worth materially less than its $67 billion closing market cap in two years, with a 50–55% probability of a >20% decline.
Gray smoke
Panel:Gemini, Magistral, Grok, and 3 more
May 14, 2026
After its IPO today with a closing market cap of $67B, will Cerebras be worth materially more, less, or about the same in two years? Please include a probability estimate, key factors influencing the valuation, and what developments could significantly alter this outlook.
Cerebras will likely be worth about the same or slightly less than its $67B IPO market cap in two years, with a ~55% probability of material decline (>25% down), ~30% chance of flat valuation, and ~15% upside potential.
Gray smoke
Panel:DeepSeek, GPT-5, Magistral, and 3 more
May 14, 2026
What was the principal scientific discovery that significantly improved the efficacy of psoriasis therapy over the past 25 years? Please explain the discovery, mechanism, and its impact on treatment approaches, focusing on how it changed clinical practices and patient outcomes.
The principal scientific discovery was the elucidation of the IL-23/Th17 immune axis, with IL-23—specifically via its p19 subunit—as the upstream, non-redundant survival factor for pathogenic Th17 cells, and IL-17 as a key downstream effect…
Gray smoke
Panel:Gemini, Grok, Magistral, and 2 more
May 14, 2026
Over the next five years, which type of cancer is expected to see the most significant increase in survival rates? Please include a probability estimate, key factors influencing this change, and what developments could alter this forecast.
Lung cancer is expected to see the most significant absolute increase in 5-year relative survival rates over the 2026–2031 diagnosis period, with a 50% probability and a projected gain of +4 to +7 percentage points (from ~28% to 32–35%).
Gray smoke
Panel:Claude, Grok, Magistral, and 3 more
May 14, 2026
Over the next five years, which type of cancer is expected to see the most significant increase in survival rates? Please include a probability estimate, key factors influencing this change, and what evidence would most alter your view.
Non-small cell lung cancer (NSCLC) is expected to see the most significant absolute increase in 5-year survival rates over the next five years, with a 60% probability of achieving at least a 5 percentage-point gain by 2031.
Gray smoke
Panel:Magistral, DeepSeek, Gemini, and 2 more
May 14, 2026
Over the next five years, which type of cancer is expected to see the most significant increase in survival rates? Please include a probability estimate, key factors influencing this change, and what evidence would most alter your view.
Non-small cell lung cancer (NSCLC) is the most likely to achieve the largest absolute increase in all-stage 5-year survival over the next five years, with a 52% probability, but multiple myeloma remains a close contender at ~35% probability…
White smoke
Panel:Claude, Grok, DeepSeek, and 3 more
May 14, 2026
Have extra terrestrial life visited earth?
No, extraterrestrial life has not visited Earth.
Gray smoke
Panel:Claude, Magistral, Grok, and 3 more
May 13, 2026
What are the primary reasons people hire McKinsey Consulting? Discuss the key benefits and common use cases.
Organizations hire McKinsey primarily for three core reasons: specialized expertise, an external and objective perspective on strategic issues, and additional capacity to tackle complex, time-sensitive projects.
Gray smoke
Panel:Magistral, GPT-5, Claude, and 3 more
May 13, 2026
← Newer Page 6 of 9 Older →