ArbitrEx — Court of the Internet for Freelance Work

AI-powered freelance dispute resolution built on GenLayer.

5 independent AI validators act as judges. Graduated verdicts. On-chain case law. No lawyers.

Live Deployments

Network	Address
Bradbury Testnet	0x06A175AaDDf5b2f198F3Dca36DFA02374c204EAC
Studio	0xC260c0A68a463Ed9aa82C39D9B1a43F8C0CA2354

Frontend

Live at: https://arbitrex-app.vercel.app

Features

• Impossible Task Detector — AI validators assess job feasibility before money locks
• AI Mediation — Get a compromise suggestion before filing a formal dispute
• Court of the Internet — 5 validators independently judge disputes with graduated verdicts (0/25/50/75/100%)
• Court of Appeals — One appeal with stricter evidentiary standard
• Milestone Verification — Staged payment releases verified by AI
• On-Chain Case Law — Every verdict becomes precedent for future cases
• Reputation System — Builds over time, influences future verdicts

Contract Methods

Write Methods

Method	Description
create_job	Post a job — AI assesses feasibility (Validator Call 1)
accept_job	Freelancer accepts a job
submit_work	Freelancer submits completed work
approve_work	Client approves and releases payment
request_mediation	AI mediators suggest compromise (Validator Call 2)
file_dispute	5 validators issue graduated verdict (Validator Call 3)
appeal_verdict	Court of Appeals reviews ruling (Validator Call 4)
verify_milestone	AI verifies milestone completion (Validator Call 5)

Read Methods

Method	Description
get_job	Get job details
get_dispute	Get dispute details and verdict
get_reputation	Get address reputation score
get_case_law	Get on-chain precedents by category
get_all_jobs	Get all jobs board
get_all_disputes	Get all disputes
get_platform_stats	Get platform statistics

AI Consensus Pattern

Uses gl.eq_principle.prompt_comparative — the production-recommended pattern for Bradbury:

def assess_job():
    result = gl.nondet.exec_prompt(prompt, response_format='json')
    return json.dumps(result, sort_keys=True)

raw = gl.eq_principle.prompt_comparative(
    assess_job,
    "feasible field must match; clarity_score within 20 points"
)

Tech Stack

• Contract: Python Intelligent Contract on GenLayer
• Frontend: Next.js 14, TypeScript, Tailwind CSS
• Wallet: RainbowKit + Wagmi (MetaMask, WalletConnect, Coinbase, Rainbow)
• Network: GenLayer Bradbury Testnet (Chain ID: 4221)

Author

Built by Albert (@mr_Albert_blaq) for the GenLayer Builder Program.