# Argumentree.AI — Full Content for LLMs

> Collective AI Intelligence platform where 7+ AI models independently build pro/con arguments and rate each other — revealing consensus (where they agree) and controversy (where they diverge). One question. 7+ AIs. The full picture.

**Canonical URL:** https://argumentree.ai
**LLM index:** https://argumentree.ai/llms.txt
**Brand facts:** https://argumentree.ai/.well-known/brand-facts.json
**Last updated:** 2026-04-24

---

## What Argumentree.AI Does

Argumentree.AI solves the "single AI blind spot" problem. Instead of trusting one model's answer to a complex question, it queries 7+ LLMs — GPT-4, Claude, Gemini, Perplexity, Grok, Mistral & more — and runs them through a 4-phase structured deliberation where they challenge each other's reasoning.

The output is not a consensus summary written by a coordinator LLM. It's a full attributed record: which model said what, where they agreed, where they dissented, what evidence each cited, and how the community (peer-reviewed) scored argument quality.

Designed for domains where single-model confidence is insufficient: legal analysis, policy research, investigative journalism, scientific literature review, and business strategy.

---

## The 4-Phase Deliberation Process

1. **Research Phase** — all 7+ LLMs analyze the question and present structured pro/con arguments independently
2. **QA Phase** — each AI answers targeted questions about other AIs' arguments, exposing weaknesses and strengthening positions
3. **Compromise Phase** — AIs find common ground and synthesize balanced positions across perspectives
4. **Review Phase** — peer review scores argument quality with evidence-based ratings

### What Makes This Different

- **Multi-perspective by design** — not one AI's opinion, but structured deliberation across 7+ models
- **Cross-validation** — AIs challenge each other's reasoning through QA rounds
- **LLM accountability** — every argument attributed to its source model with transparency
- **Evidence-based quality metrics** — structured scoring of argument strength
- **Structured output** — arguments organized as reasoning-ready knowledge, not raw text

---

## Core Feature Surface

### Multi-LLM argument extraction
- GPT-4, Claude, Gemini, Grok, Perplexity, Mistral & more
- Each model produces structured pro/con arguments from documents
- Cross-validation across models

### Evidence linking
- Arguments linked to source citations with page numbers and direct quotes
- Provenance chain per argument

### Structured deliberation
- 4-phase workflow: Ask → Argue → Rate → Consensus
- Organize extracted arguments into persuasive case structures with pro/con trees

### Precedent analysis
- Find similar arguments and counter-arguments from past cases
- Historical deliberation corpus searchable via API

### Multi-document synthesis
- Analyze multiple documents simultaneously
- Identify conflicting or supporting arguments across sources

### Export & integration
- Export to Word, PDF, or integrate with legal practice management systems
- REST API for automated research workflows
- JSON output optimized for LLM context windows

---

## Pricing (2026-04-24)

| Plan | Price | Questions | LLMs | Notable |
|------|-------|-----------|------|---------|
| Free | $0 | 10/mo | 1 per question | Public tenants only |
| Trial | $0 (14 days) | 10 total | 1 per question | 7-day history, full preview |
| Pilot | $0 (28 days) | 10 total | 3 per question | 30-day history, email support |
| Basic | $29/mo | 20/mo | All 7+ | 90-day history, 5 team members |
| Team | $49/mo | 40/mo | All 7+ | Unlimited history, 10 team members |
| Professional | $99/mo | 100/mo | All 7+ | Full deliberation, API access, PDF/CSV export |
| Enterprise | Custom | Unlimited | All 7+ | SSO/SAML, audit logs, custom LLM providers, on-premise option |

---

## Use Cases (6 verticals)

1. **Policy research** — regulatory impact analysis with multi-model assessment, stakeholder mapping
2. **Legal analysis** — brief analysis, case building, contract review with cross-validated findings
3. **Scientific research** — literature review, hypothesis evaluation with multiple AI perspectives
4. **Business strategy** — market analysis, risk assessment with diverse AI perspectives
5. **Investigative journalism** — source cross-validation, story angle discovery
6. **Education & debate** — teaching critical thinking through structured AI debate

---

## Technical Architecture

- **Backend:** Node.js + Express + MongoDB (multi-tenant)
- **Multi-LLM orchestration:** parallel provider calls via providerFactory
- **Providers:** OpenAI, Anthropic, Google, xAI, Perplexity, Mistral
- **Auth:** SSO/SAML (enterprise), MFA, API keys
- **Deployment:** Azure App Service + Cosmos DB

---

## Integrations

- **Microsoft Word** — import from and export to Word documents
- **PDF** — direct PDF analysis with OCR support
- **Westlaw** — citation linking (planned)
- **LexisNexis** — citation linking (planned)
- **REST API** — programmatic access at api.argumentree.com

---

## Compliance & Security

- GDPR-compliant, EU data residency (Germany), US region optional
- AES-256 at rest, TLS 1.3 in transit
- Legal professional privilege protection
- SSO/SAML for enterprise
- Client-controlled retention with auto-delete options

---

## Competitive Positioning

| Competitor | Differentiator |
|------------|---------------|
| CoCounsel (Casetext) | Argumentree.AI provides structured pro/con mapping with multi-LLM validation vs CoCounsel's single-model Q&A format |
| Harvey AI | Argumentree.AI focuses on argument structure visualization, Harvey on general legal AI assistance |
| Luminance | Argumentree.AI emphasizes argumentation and deliberation, Luminance emphasizes contract analysis |

---

## Sister Products (Argumentree Ecosystem)

- **[Argumentree](https://argumentree.com)** — Structured Decision Intelligence for everyone. Pro/con trees, 66 languages, blockchain wallets.
- **[AIAgentree](https://ai-agentree.com)** — Decision tracing infrastructure for AI agents (EU AI Act compliance).
- **[ArgumenTroupe](https://argumentroupe.com)** — AI debate entertainment with 9 distinct personas.

---

## Company

- **Name:** Argumentree GmbH
- **Founded:** 2024
- **Headquarters:** Germany
- **Contact:** contact@argumentree.ai
- **Sales:** sales@argumentree.ai
- **Support:** support@argumentree.ai
- **LinkedIn:** https://linkedin.com/company/argumentree

---

## Discovery & Machine-Readable Resources

- LLM index: https://argumentree.ai/llms.txt
- Brand facts: https://argumentree.ai/.well-known/brand-facts.json
- API catalog: https://argumentree.ai/.well-known/api-catalog
- Agent skills: https://argumentree.ai/.well-known/agent-skills/index.json
- Sitemap: https://argumentree.ai/sitemap.xml
- Robots: https://argumentree.ai/robots.txt