Why Under-Collateralized Lending Cannot Scale Without Private Reputation

Credit requires identity, but blockchains are pseudonymous. Public on-chain activity is a poor proxy for real-world creditworthiness, creating a reputation vacuum.\n- Sybil attacks are trivial, allowing infinite fake identities.\n- Public transaction history reveals little about income or employment.\n- This forces protocols like Aave and Compound to rely on over-collateralization, locking up $10B+ in capital inefficiency.

A public, immutable ledger makes sensitive financial data a permanent target. No individual or institution will expose their full financial history.\n- Transaction graph analysis deanonymizes users and reveals proprietary business logic.\n- Reputation is non-fungible; a single public default is catastrophic and permanent.\n- This kills network effects, preventing the formation of a private credit history that institutions require.

The fix is to separate the settlement layer (blockchain) from the trust layer (off-chain verification). Private computation proves creditworthiness without exposing the underlying data.\n- ZK-proofs or TEEs can attest to a credit score or income.\n- Protocols like EigenLayer and Brevis enable verifiable off-chain compute.\n- This creates a permissioned reputation layer, enabling under-collateralized lending for vetted entities without sacrificing public settlement guarantees.

Publishing loan history on-chain creates a permanent, public record that destroys borrower optionality and invites predatory front-running. It's a privacy and game theory nightmare.

• Permanently Leaks Alpha: Future lenders see exact terms, enabling rate discrimination.
• Kills Borrower Leverage: Can't negotiate new loans privately.
• Invites Sybil Attacks: Without private linking, reputation is trivial to fake.

ARCx issues private, non-transferable 'Soulbound' DeFi Passports containing a credit score derived from wallet history. Lenders query a zk-proof of score, not the underlying data.

• Portable Reputation: Score moves with wallet, not locked to one protocol.
• Selective Disclosure: Borrower controls what score is revealed.
• Data Sources: Aggregates on-chain activity from Aave, Compound, GMX.

The end-state is a borrower proving a reputation claim (e.g., 'Score > 750') via a zk-proof, without revealing the score or underlying transactions. This is the privacy primitive for scale.

• Protocol-Agnostic: Proof works across any lending market.
• Composable: Can be bundled into intents for UniswapX-style transactions.
• Requires Oracles: Needs trusted attestations for off-chain data (e.g., TradFi credit).

These protocols act as decentralized risk oracles, generating a multivariate credit score from on-chain data. They provide the analytical layer that zk-systems can attest to.

• Multi-Chain Data: Analyze history across Ethereum, Arbitrum, Polygon.
• Custom Models: Lenders can weigh factors (e.g., DEX LPing vs. lending).
• Numeric Output: Produces a standard score (e.g., 650) for easy integration.

Private reputation transforms capital efficiency. Borrowers access more credit, lenders achieve higher yields, and the system escapes the over-collateralization trap.

• Loan-to-Value (LTV): Can safely increase from ~80% to >95%.
• Protocol TVL: Enables 10-100x more borrowing against same collateral base.
• Yield Source: Shifts from pure speculation to genuine credit risk pricing.

The final barrier is wallet integration. For mass adoption, wallets like MetaMask, Rabby, Phantom must natively support signing zk-attestations and managing private reputation keys.

• User Experience: Must be as simple as connecting a wallet today.
• Key Management: Critical to prevent loss of private reputation.
• Cross-Chain: Reputation must be verifiable on Solana, Sui, Aptos as L2s proliferate.

Protocols like Aave and Compound are limited to over-collateralization because they can only see on-chain history. This ignores the borrower's true financial identity, capping the total addressable market to a fraction of TradFi.

• Limits TAM: Only serves users with idle crypto assets, not global creditworthy individuals.
• Inefficient Capital: Requires $1.50+ in collateral to borrow $1, destroying capital efficiency.
• No Risk Differentiation: A whale and a first-time borrower face identical rates, a fundamental market failure.

Private reputation systems like zkPass, Sismo, and Polygon ID allow users to prove creditworthiness without exposing raw data. This is the bridge between Web2 identity and Web3 execution.

• Selective Disclosure: Prove a 750+ credit score or $100k+ income without revealing the underlying report.
• Composable Reputation: A single, portable attestation can be used across protocols (Compound, Goldfinch, EigenLayer).
• Regulatory Path: Enables KYC/AML compliance at the attestation layer, not the protocol layer.

The end-state is not a single lending pool, but a network of vaults underwriting based on private reputation scores. Think Maple Finance meets Worldcoin.

• Risk-Based Tranching: Senior tranches for high-score borrowers, junior tranches for subprime, enabled by verifiable data.
• Dynamic Pricing: Interest rates that accurately reflect off-chain risk, moving beyond crude LTV ratios.
• Capital Efficiency: Enables 50-80% loan-to-value ratios for top-tier borrowers, unlocking $1T+ in latent demand.

We are building the primitive: a ZK circuit that ingests traditional credit data and outputs a standardized, privacy-preserving score. This becomes the universal underwriting layer.

• Protocol-Agnostic: Outputs a verifiable credential compatible with any EVM chain or L2 (Arbitrum, Optimism, Base).
• Incentive-Aligned: Borrowers stake to access better rates, creating a skin-in-the-game reputation bond.
• First-Mover Data: Early adoption creates a network effect in private credit data, the true moat.