Summary: Chaos Labs is a risk-and-data operator for crypto markets. The interesting shift is that it no longer stops at advisory work: parameter guidance, monitoring, machine-consumed feeds, attestations, and AI-facing market analysis now sit in one stack, which makes it closer to a control plane than a consulting memo factory.
What it does:
Provides risk-management and analytics services for DeFi protocols, including parameter recommendations and monitoring around collateral, solvency, liquidations, and market volatility
Operates oracle-style data products that expose pull/push price feeds, protocol risk feeds, and proof/attestation feeds in official product materials
Markets an AI product that answers market questions using Chaos Labs’ proprietary crypto datasets and research infrastructure
Works with DeFi protocols, crypto-native organizations, and digital-asset institutions rather than retail end users as its primary customer base
Maintains a public GitHub organization with oracle-testing utilities, price-feed tooling, and agent-related/open-source repos that complement its research-and-infrastructure positioning
Key claims:
Homepage says strong and proactive risk management helps protect protocols against bad debt, depegging, cascading liquidations, oracle manipulation, and other market risks
Official site says Chaos Labs serves leading DeFi protocols, crypto-native organizations, and select digital-asset institutions, naming customers such as Aave, Jupiter, and GMX
Oracles product page shows Chaos Labs presenting pull/push price feeds, protocol risk feeds, and proof feeds such as Ethena USDe attestations as part of one risk-data surface
AI product page says Chaos AI is built on years of proprietary data from securing over $200 billion in crypto assets and is intended to deliver “hedge fund-quality analysis” in seconds
About page frames the company around real-time financial intelligence, citing large cumulative transaction-settlement, volume, and value-secured figures as proof of operational scale
Whitepaper: No classic whitepaper or litepaper was found during this pass. The strongest primary materials were Chaos Labs’ homepage, about page, oracles product page, AI product page, AI docs landing page, and public GitHub organization; see ../whitepapers/chaos-labs-primary-sources-2026-04-25.md.