Chaos Labs Risk Platform Proposal
01Summary
Aave DAO approves a 12-month engagement with Chaos Labs to build risk and simulation tooling for Aave v3 markets. Payment of 500,000 aUSDC ($0.5M) streams over 6 months starting 6 months after approval, with potential additional incentive-based payments. The DAO can terminate after 6 months before payment begins.
- Type
- treasury allocation
- Funding
- $500000
Key changes
- Establishes 6-month payment stream of 500,000 aUSDC ($0.5M) to Chaos Labs starting 6 months after on-chain approval(high)
- Chaos Labs will provide risk management services including borrow/lending caps recommendations, collateral factor recommendations, interest rate recommendations, asset listing risk analytics, and community risk calls for all Aave v3 markets(high)
- DAO retains right to terminate engagement after 6 months prior to payment stream initialization(medium)
- Additional incentive-based payments to be proposed separately after service delivery(medium)
Notable concerns
- Payment delayed 6 months from approval before streams commence
- Additional incentive-based compensation amounts unspecified and subject to future proposals
- Engagement period is 12 months total but only 6 months of specified payment detailed
Generated by Kvorum
02Description
Simple Summary
This proposal outlines a 12-month engagement for Chaos Labs to build and utilize risk and simulation tooling to help optimizations and protect Aave v3 markets.
Abstract
This proposal acts as a mandate from the Aave community to engage with Chaos for the work defined on the Aave governance forum here and pre-approved by the Aave community via Snapshot here.
Motivation
Chaos Labs is eager to collaborate with Aave to ensure it maintains its position as a best-in-class DeFi protocol with one of the most serious and intentional risk operations. Aave v3 has over $1b in locked TVL and will shortly be launching on Ethereum, drastically increasing the assets controlled and the users engaging with it. Chaos has previously built a v3 Risk Dashboard to help monitor these positions and is excited to expand that capability using its proprietary on-chain simulation engine.
Chaos Labs will focus on risk management and parameter recommendations for all v3 markets.
This includes:
- Borrow & lending caps recommendations
- Collateral factors recommendations
- Interest rates recommendations
- Asset listing risk analytics & tooling
- Open-sourced agent access
- Extending the AAVE v3 Risk Analytics App
- Community Risk Calls
While incorporating feedback from the community, we have committed to building trust between the Aave community and Chaos Labs. To this end, the payment streams will not start until 6 months after AIP approval with two incentive-based payments not included in this proposal but will be submitted after delivery. The DAO may terminate our engagement after 6 months prior to this stream initializing.
Specification
A full specification of the payload contract and tests enabling this mandate can be found on the Chaos Labs Github here.
In summary, the proposal creates a 6-month stream of 500,000 aUSDC ($0.5 Million) to the Chaos-controlled address starting at block height equivalent to the 6-month anniversary of on-chain approval.
Test Cases
The Proposal Payload has been tested and peer-reviewed by Bored Ghost Developing, including simulations on mainnet of the whole proposal lifecycle.
Link to Test Cases: Here
Implementation
Link to Payload Implementation: Here
Deployed Contracts ProposalPayload = 0x691B41805f7Ef2D7De6165bC42295b035a31600D
Copyright
Copyright and related rights waived via CC0.
03Mismatch analysis
The proposal follows the standard Aave governance pattern: a single top-level action invoking execute() on a pre-deployed, audited ProposalPayload contract, whose address (0x691b41805f7ef2d7de6165bc42295b035a31600d) matches exactly the address cited in the description's 'Deployed Contracts' section. The actual mechanics described in the prose (creating a 500,000 aUSDC stream to Chaos Labs starting 6 months post-approval) are implemented inside that payload contract's internal logic, which is not decodable from this single execute() call — this is normal and expected for Aave AIP payloads that were tested/peer-reviewed (as noted, tested by BGD Labs) rather than a red flag. Since the on-chain calldata target matches the disclosed payload address and no additional or missing top-level actions are present, there is no evidence of a target mismatch, omitted action, or value discrepancy at the level visible in this data. Confidence is medium rather than high because the specific numeric details (500,000 aUSDC amount, recipient address, 6-month timing) cannot be independently verified from the flat calldata alone — verifying those would require inspecting the payload contract's bytecode/source, which is referenced but not included in the decoded actions provided.
Generated by Kvorum
04Actions
- Target
- 0x691b…600d
- Value
- 0 wei
- Function
- fallback()
- Arguments
{}- Calldata
05Tally
Summed from votes
06Voters
No votes recorded yet.
07Forum
- Updated Proposal: Chaos Labs - Risk & Simulation Platform ↗discourse · high confidence
Arguments for
- Chaos Labs' simulation platform provides valuable risk analysis and parameter recommendation tools that complement existing risk vendors and introduce healthy competition to improve overall risk management quality.— lucasvo, DiamondRock, fig, PennBlockchain, rspa
- The revised proposal with zero upfront cost for 6 months and the ability to terminate without payment significantly reduces financial risk to the DAO while allowing Chaos to prove its capabilities.— omergoldberg, fig, PennBlockchain
- Chaos Labs has demonstrated strong prior work for Aave including the v3 Risk Application and stETH depeg analysis, and has a credible team with experience from major technology companies.— omergoldberg, DiamondRock, PennBlockchain, rspa
- Multiple risk vendors with different methodologies create beneficial redundancy and diverse perspectives on risk parameters, reducing centralization risk and vendor lock-in common in traditional finance.— omergoldberg, DiamondRock, lucasvo, fig
- The platform's open-sourcing of agents and transparency-focused tools will democratize risk analysis and empower the broader community to conduct independent risk assessments.— omergoldberg, lucasvo
- Aave's growth to 10x scale requires additional scalable risk management infrastructure and tools beyond current vendors.— fig, omergoldberg, PennBlockchain
Arguments against
- Chaos Labs' case studies lack statistical rigor, make unrealistic assumptions, and demonstrate fundamental misunderstandings of agent-based modeling and liquidation parameter optimization needed for billions in AUM.— tarun
- Multiple risk managers making competing parameter recommendations creates governance confusion and accountability gaps if one vendor's recommendations cause protocol losses.— inkyamze, eboado
- Chaos Labs has no track record of governance proposals or proven ability to calibrate models to historical events, unlike Gauntlet and other established risk vendors.— tarun, inkyamze
- Chaos Labs should seek funding through the Aave Grants DAO rather than pursuing a direct engagement, which would be more appropriate for an early-stage project.— eboado
Unresolved concerns
- Whether Chaos Labs' open-sourced agents and models will actually provide sufficient transparency for community verification of risk parameter recommendations.— tarun, eboado
- How the DAO will handle conflicting parameter recommendations if both Chaos Labs and Gauntlet submit competing proposals.— inkyamze, eboado
- Whether Chaos Labs conducted adequate due diligence on existing Aave risk management procedures and community requirements before proposing its engagement.— eboado
- Why Chaos Labs' simulations did not identify and flag attack vectors similar to those used in the Mango Markets hack.— Jommi
Notable participants
- omergoldbergCEO of Chaos Labs; primary proposal author who defended the team against technical criticisms and revised pricing to zero upfront cost for 6 months.
- tarunGauntlet researcher who provided detailed technical critique of Chaos Labs' statistical rigor, modeling methodology, and case studies.
- inkyamzeGauntlet representative who articulated concerns about multiple risk managers and proposed Chaos remove parameter recommendations to gain support.
- lucasvoEarly supporter who argued Chaos Labs' tools are complementary to Gauntlet and welcomed open-sourced transparency improvements.
- DiamondRockBenqi CEO providing third-party testimony of Chaos Labs' quality work and positive track record with another lending protocol.
- figCommunity member who challenged bias against new vendors, advocated for openness to innovation, and supported Chaos Labs despite lacking complete technical understanding.
- eboadoCritical community voice who questioned vendor lock-in risks, noted insufficient due diligence, and opposed the free 6-month model as creating soft obligations.
- PennBlockchainDelegate who voted YES based on Chaos Labs' prior work quality, the 6-month free period, and the ability to off-board with minimal cost.
- rspaFlipside Governance delegate who reported positive impression from prior Chaos Labs interactions and supported the proposal based on demonstrated competence.
- AaveLabsOfficial Aave Labs representative offering cautious support for the proposal while emphasizing the importance of strong risk management.
Generated by Kvorum
08Similar proposals
- aaveexecutedChaos Labs Aave v2 Coverage90%
- aaveexecutedChaos Labs Scope and Compensation Amendment86%
- aaveexecutedChaos Labs Risk Management Renewal86%
- aaveexecutedChaos Labs <> Aave Risk Management Service Renewal85%
- aaveexecutedChaos Labs Engagement Amendment85%
- aavecanceledChaos Labs <> Aave Risk Management Service Renewal85%
- aaveexecutedChaos Labs x Aave DAO — Early Renewal Proposal82%
- aaveexecutedChaos Labs - Payment Collection Request80%
- aaveexecutedOrderly Transition and Offboarding Plan for Chaos Labs79%
- aaveexecutedTokenLogic & karpatkey Service Provider Partnership75%