[ARFC] Gauntlet <> Aave Renewal 2023

I really appreciate Gauntlet is sticking with the DAO for another year to manage risk together with @ChaosLabs. I think both have been working together in a great manner providing important information.
Thus i support this engagement and hope to see another great year of cooperation.

Just one question, would you be up to change the payment from 40% GHO to 60% and thus minimize aUSDC to 40% from 60%?
This would show strong support to the DAO and GHO but also to your engagement regarding GHO and its peg. (more skin in the game, therefor giving 100% towards GHO)

Thank you

5 Likes

We’re supportive of Gauntlet’s renewal for another year as we believe risk service providers are essential to the livelihood of Aave.

Just a note on showing support for GHO by receiving 100% streams in GHO.
We have to be realistic here. While it’s great for all teams to be paid in GHO and we’re supportive of this - risk service providers that receive a sizeable amount of streams, need to pay their team members. For that to happen, it means selling GHO in the market for USDC. This would be counterproductive on the DAOs efforts to bring GHO back to peg.

So it makes sense that a healthy mix of both GHO and aUSDC is proposed here. Chaos Labs proposed 50-50 and it’s now on AIP stage. We don’t see any issue with 40-60 as it currently stands but a good middle ground would be 50-50.

If we’re talking about alignment and skin in the game, this is great:

4 Likes

Blockchain at Columbia is in support of this proposal. Gauntlet plays a pivotal role in ensuring the stability of AAVE, backed by a remarkable track record of exceptional risk management—a rarity in the industry. Furthermore, Gauntlet’s commitment to expanding its scope without raising fees reinforces our endorsement of this initiative

3 Likes

Index Coop supports this proposal and we are happy to see Gauntlet continue to work with the Aave DAO. Since the beginning of DeFi, Gauntlet has set the standard for risk management. Over the past few years, Gauntlet’s simulations and recommendations (which have sometimes been ignored) have often proven to be correct and always in the best interest of AAVE holders.

Our team believes that GHO is critical for the continued success of Aave and we are happy to see the expanded scope in that area as well as payment in GHO. Furthermore, no other service providers have baked-in refunds and clearly shows why Gauntlet is superior to other providers.

4 Likes

We are in support of Gauntlet’s continued engagement with the Aave DAO. Gauntlet has proved to be invaluable to the Aave DAO, and we would be glad to see Gauntlet continue the great work.

We are particularly impressed by the insolvency refund, the refund is a clear example of what being truly aligned means.

7 Likes

Hi @Gauntlet,

Thank you for all the contributions Gauntlet has made to the Aave DAO over the past year. The inclusion of GHO in the stream outside of the insolvency fund was a generous gesture that is greatly appreciated.

Upon reviewing the proposal, I believe there is room for improvement. The below details three key areas where I think Gauntlet needs to improve:

  1. Response Time and Availability

Last year, Gauntlet committed to enhancing communication across the DAO, and I think there is still more that can be done in this area. During the period of frequent updates to Supply Caps, communication was slow, with a standard response time was approaching a week. Meanwhile other teams were able to turn around analysis a lot faster.

Even now, Gauntlet is a Monday to Friday business when DeFi is 24/7. Teams like the ACI, Chaos Labs, BGD and TokenLogic are available 24/7. General responsiveness needs to improve and there is no reason why the Risk Stewards are unable to push parameter changes on the weekend if caps are full.

The Supply Cap situation resulted in stalled growth on Polygon, causing frustration among partners providing Liquidity Mining (LM) incentives and communities who built products on Aave. Some patterns were going to do LM across several Aave v3 deployments and due to the Supply Cap issue on Polygon, were deterred and this hurt Aave’s growth prospects. The builders who developed structured products on Aave v3 experienced stalled launch programs and faced frustrated users unable to access the automate incentivised yield-maximizing strategies.

  1. Product/Dashboard

Gauntlet has invested heavily in modelling Aave and the broader ecosystem. Whilst Gauntlet’s dashboards are insightful, there is room for improvement. Looking at the Gauntlet Dashboard, the risk parameters settings for each Aave reserve are not to be found. Furthermore, there is no analysis indicating what the parameter configuration should be given prevailing market conditions.

There are many other examples that can be drawn by comparing the Chaos Labs and Gauntlet dashboard to highlight where Gauntlet can improve.

  1. Pricing

Please explain why the premium relative to Chaos Labs ?

While I understand that the 2023 contract was a $2M per year commitment, I believe the 2024 commitment should align with Chaos Labs at $1.6M. My personal opinion is Chaos Labs delivery and engagement with the community is a level above Gauntlets and I don’t see evidence of the Gauntlet premium in terms of impact at Aave.

A clear example, Chaos Labs shared insights into their proposal and sought feedback from TokenLogic. Gauntlet did not seek feedback from TokenLogic despite open dialogue relating to the composition of funding.

In addition to the points above, can Gauntlet provide insight into the decision to provide its mechanism design service to aspiring liquidity protocols, like Morpho ?

Given Gauntlet’s level of involvement at both Compound and Aave, how does Gauntlet intend to ensure the learnings from Aave and Compound are not directly incorporated into Morpho v2 ? This appears to be a large portion of the value additive service offered by Gauntlet whilst not in the interest of either Aave or Compound.

Vote: NAY

I am open to amending how we intend to vote provided the price is revised and the proposal is updated to incorporate the broader feedback.

Author: @MatthewGraham

5 Likes

Hi @MatthewGraham - thank you for your thoughts. We appreciate the feedback and would also like to clarify some misperceptions.

Gauntlet’s risk monitoring platform is 24/7. There are many examples of Gauntlet working around the clock to protect Aave, including market events, USDC Non-Parity Event, and most recently the security incident.

We think you may be referring to Gauntlet not being available 24/7 to accommodate all of TokenLogic’s requests. We agree that we should have been faster with cap updates at the beginning of the year. We believe this has since improved, and the Risk Steward is being used much more frequently now. We commit to doing better here, and if the cap is indeed full, we of course commit to using the Risk Steward regardless of what day it is. We are happy to set up a recurring meeting with you to ensure we are prioritizing your requests commensurately along with the critical risk management work we do to ensure Aave is safe.

We’d note to the community that different service providers focus on different areas to serve the DAO. TokenLogic has focused on Growth among other things. We have disagreed with TokenLogic many times, including increasing caps for MaticX. The priorities of other service providers may not always line up with the priorities of our priorities, and we usually aim to prioritize workstreams related to Risk when it comes to speed. When it comes to requests, we may at times prioritize quality over speed in order to uncover potential hidden risks. As Risk Managers, our commitment is foremost to the DAO. We look to collaborate in a professional capacity with any service provider the DAO chooses. We are here to provide objective risk analysis, even when others disagree with us given differing objectives, and we are aligned by having skin in the game.

Everyone is certainly entitled to their opinion of the Dashboard. Usually community members request user studies, so if this is of interest, it would be great to discuss in more detail.

We are focusing on the highest impact services. Dashboards, from all the users studies we have performed, rank below other services including GHO Peg analysis, LST Killswitch, and DeFi tail-event research. That said, we have made significant improvements - please see our account explorer.

To be abundantly clear, Morpho was a one-time economic audit, similar to how smart contract auditors conduct audits across protocols. We do not currently serve Morpho.

Earlier this year we launched Gauntlet Applied Research. This team performs R&D often on an ad-hoc, short-term basis. While it may be perceived Morpho is competitive, we are very cognizant of the strategic implications this would have for Aave, and under no circumstances would GAR take on work with unacceptable conflicts of interest. While we don’t view any conflict of interest, we commit to 100% transparency and will let the DAO know if we ever do serve Morpho.

At a high level, the research and modeling work we do improves the value of our platform overall for all clients as we iterate on assumptions around factors like liquidity elasticity and liquidator behavior.

On pricing: to be clear, your guidance on a fixed $2M was a key input last year so that we can give the DAO predictability. While you are mentioning a budget of $1.6M now, less than a month ago you mentioned a combined Risk budget of $2.5M (implying $1.25M for Gauntlet).

We aim to have a very principled-based approach towards our pricing structure. TLDR; we are expanding scope without increasing our fee. This results in predictability for the DAO and Gauntlet.

We appreciate your comment around GHO payment and we always aim to be fully aligned with the DAO.

We are open to being flexible here @MatthewGraham @EzR3aL @0xkeyrock.eth. As such, we will increase our GHO portion to 50%. The rest will be taken in aUSDT (given the relatively lower amount of aUSDC reserves).

Summary

  • At the end of the day, if the community would like to continue Gauntlet’s full-suite risk management services with the DAO, we would be excited to have a proposal that makes sense for the DAO.
  • To clarify - our foremost commitment is to the DAO. We are open to working with any service provider the DAO chooses and commit to collaborating more without losing sight that our primary function is Risk Management.
9 Likes

One of the things that we agree is the pricing. Both @Gauntlet and @ChaosLabs should have a similar budget, or even equal. I don’t understand what kind of things Gauntlet is offering that Chaos is not. I also agree that Chaos has a way better way of communicating with the DAO than Gauntlet. There is a lot of room for improvement. I will vote for NAY (for now)

1 Like

Franklin DAO stands with Gauntlet’s efforts to continue contributing to Aave’s ecosystem. Their track record and commitment to protocol alignment resounds well with our core initiatives of promoting sustainable growth.

Gauntlet has always been prompt to report issues related to protocol risk as well as protocol improvements. This is present on a 24/7 basis. We greatly respect the continued effort to notify the community of risks as well as potential upside. The upside in this newest proposal relates to the support of GHO stability which we strictly requested.

All in all we stand by gauntlet’s involvement with the future of Aave and their collaborative efforts with the community to sustain the protocol.

5 Likes

The ACI does not support the renewal of Gauntlet’s engagement.

We recall that last year’s renewal proposal was controversial, and despite this, we decided to support Gauntlet, believing the benefits of having a risk service provider and redundancy outweighed the downsides and costs.

We also remember the proposal that craftily aimed to extract millions from the Aave DAO, which is one of the reasons I created the ACI. This behavior no longer exists because service providers have since switched to propose fair pricing, or the DAO has ceased working with them.

The DAO is now in a more robust and stable position, and with more experience, we can take a step back and reflect on the benefits of working with Gauntlet. With the ACI, we are now unsupportive of renewing our collaboration with Gauntlet, and here’s some of the reasons why:

DeFi is a 24/7 endeavor, and working for a DAO is like working at a startup: it’s exciting, it’s a learning experience, and it can bring good ROI in terms of brand & reputation, but it’s also tireless and stressful.

Most Aave DAO service providers know what it means to be on the frontlines & sitting in a war room. Very recently, with the responsible disclosure event, BgdLabs led the coordination that resulted, 9 days later, in a zero user fund loss & back to normal outcome.

What was Gauntlet’s contribution to this? They provided an analysis on the public forum about how the pause could create bad debt in Aave. This kind of analysis is valuable for making informed decisions in a war room context and should be published in a post-mortem disclosure after the events. Publishing this in a public forum during events could have created more panic than necessary and is a behavior we consider unfit and lacking strategic vision.

Gauntlet is also by far the slowest service provider.

For example, the Risk Steward was introduced to allow for a fast-track, governance-less process to act on supply & borrow caps. It’s an essential tool to ensure growth is not artificially limited. However, due to Gauntlet’s poor reactivity, the Risk Steward has been a semi-failure. In this recent example on Gnosischain, it took 4 days to react to Chaos Labs’ proposition, and to our knowledge, this proposal has not been enforced yet. There are dozens of examples like this in the DAO’s history. Gauntlet simply is not fit for a 24/7, non-stop DeFi environment.

During this period, there have been many examples of Gauntlet being difficult to work with, especially with Chaos Labs. Gauntlet seems to have adopted a zero-sum game mindset with the DAO’s risk service provider. It is unfortunate because the ACI was completely open-minded to support a dual-risk service provider team to allow for redundancy & enhanced protocol safety.

With the ACI, as with every single service provider, we proposed to Gauntlet to co-author proposals with us to encourage collaboration and synergies in the DAO. We must share that working with Gauntlet was not our preferred experience, and the recent AIP-371 fiasco is a prime example of this.

Regardless of the outcome of this proposal, the ACI, except in the context of Skywards services, does not wish to collaborate with Gauntlet in the future.

For the ACI, moonlighting for direct competition is unacceptable behavior. Gauntlet accepted work for a protocol designed to compete directly with the Aave DAO.

We consider this fact alone as solid ground to launch an immediate ARFC to cut the Gauntlet stream and fire them on the spot. Without the recent responsible disclosure event and the relative near-end of their current stream that might have been our chosen path, but considering these elements, we decided to simply support the non-renewal of Gauntlet.

7 Likes

Hi @MarcZeller, thanks for your feedback. We have some thoughts to share here-

We have been key contributors to many war rooms, including the USDC depeg and the most recent security incident. We apologize that we did not include you in these war rooms - that was a miss. In addition, we have been actively informing the community with Gauntlet’s Aave market alerts telegram channel and making risk mitigation proposals for TUSD, MAI, and V2 Deprecation. 24/7 coverage of market risk is a key component of our service, and we aim to provide the highest quality and most in-depth analysis in a short time frame.

In DAOs, we recognize the importance of balancing public and private information. It is rare for everyone to always agree on what should be public and what should be private. We shared our analysis with Aave tokenholders and delegates and directly asked if they would prefer seeing this in the public forums to inform transparent decision-making. There are also additional workstreams we helped drive in war rooms, such as a detailed analysis of how long a grace period should be for the security incident.

Certainly, speed is a relative metric for the community to judge. Through high-priority market risk events like the USDC depeg, we believe we have responded quickly while maintaining the high quality and accuracy of our analysis.

Regarding Gauntlet’s joint proposal with ACI on weth interest rates, what’s most impactful to the DAO is that the changes to the interest rate for weth are now executed. Unfortunately, mistakes do happen in governance. We take ownership of the recent duplicate and are making changes in our governance process to improve and avoid duplicate AIPs going forward.

This is unfortunate. We are open to collaborating with any service provider the DAO chooses because that’s how a professional organization can function effectively. Recently, we’ve collaborated on the stablecoin interest rate proposal with Monetsupply, implementation of the Killswitch mechanism with BGD, joint v2 deprecation proposals with Chaos, and more.

Summary

As always, we value all thoughts from the community and continuously aim to provide more value to the DAO. With differing opinions in the DAO, it is up to each community member to independently judge the quality of our work. We pride ourselves on our risk output and value added to the DAO. Managing Aave’s market risk is, first and foremost, the highest of our priorities, especially during market stress events, and we will continue to focus on risk modeling and simulation going forward.

13 Likes

We are in a very fast moving industry especially when there are “war-room” type of situations but in those type of complex quantitative analysis - speed can be your enemy in getting accurate results.

In our opinion, there is always a balance between speed and quality.

9 Likes

We have found our experience with Gauntlet to be pleasant and helpful as a delegate. We’ve always received prompt communication and updates regarding their work and proposals which has been helpful. However, it is interesting to hear about TokenLogic and Marc’s experiences from a deeper operational side.

There is a clear misalignment between what growth and risk service providers optimize for, but we do agree that these misalignments should not be the reason why proposals like ones from the Risk Steward to adjust caps should be unnecessarily delayed due to waiting on a response from either risk provider.

We’d be supportive of Gauntlet’s renewal if they were to bring their compensation in line with Chaos Labs’ at $1.6M as we believe the proposed scope doesn’t warrant a $400k premium.

Overall, we value having both Chaos Labs and Gauntlet as risk providers as their combination provides differing insights and evaluations which is healthy for the DAO.

Disclaimer: Wintemute is an investor in Chaos Labs

8 Likes

[This post was written by Blockchain at Berkeley’s Governance team and represents the views of Blockchain at Berkeley]

Blockchain at Berkeley’s view on Gauntlet <> Aave Renewal 2023:

As a long-time delegate and participant in the AAVE Governance space, we would like to provide our own perspective on our experience working with Gauntlet. Throughout our activity in AAVE governance, Gauntlet has been an easy-to-work-with service provider and has willingly provided thorough explanations for their motivations behind proposals, specifically on interest rates across all Aave markets and other parameter recommendations.

We believe that Gauntlet has always provided an independent and consistent data-backed perspective even if community sentiment was the opposite (for ex, Curve risk).

Gauntlet have always taken the initiative, being the first to make recs and publish methodologies for isolation mode, stablecoin emode, LST emode on Aave V3, along with regular v2 deprecation recommendations. Gauntlet also set the standard, being the first to publish methodologies to set interest rates across all Aave markets.

As well, we have a direct communication channel with Gauntlet, who provide timely (multiple times a week) updates on the work they are doing with us and respond promptly to concerns as well. We have had a very positive experience working with Gauntlet.

6 Likes

Thank you everyone for your productive feedback. Gauntlet will be flexible to accommodate the DAO’s preferences. We plan on moving forward with a $1.6M service partnership. No meaningful scope changes will occur. Our insolvency refund scales with the notional contract size and only covers V3. Mechanism design workstreams (e.g., LST Killswitch) will continue on a best-efforts basis.

We will put up a Snapshot vote for a $1.6M partnership (20% discounted fee from our original, with the addition of taking on GHO as half of the payment) and hope that this demonstrates flexibility to the DAO while also ensuring that we can commit the resourcing to provide robust risk management services to Aave. We thank the community again for their collaboration during this process.

2 Likes

Having been intimately involved in Aave Governance for more than a year now, we can confidently say that Gauntlet has been great to work with as delegates. Their track record has been great and in our experience, they have responded appropriately to situations which require war rooming. We are in favour of renewing Gauntlet for $1.6M.

4 Likes

Thank you to all who provided their feedback. We have posted our Snapshot here and thank the community for their participation in the vote.

2 Likes

The Snapshot has passed - we greatly value all the support from the Aave community. We have published the AIP and voting begins in one day. We appreciate all the thoughts provided during this process and thank the community in advance for their participation.

1 Like

Good @Gauntlet but now try to be more proactive with defi time-shifts (24/7) :)

1 Like

This topic was automatically closed 30 days after the last reply. New replies are no longer allowed.