Add RAI pools to Tornado Cash

Hey everyone!

This post looks to kick off a conversation between the RAI and Tornado Cash communities. Myself and the Fire Eyes squad as well as the Reflexer Labs team believe that both projects are very much ideologically aligned and are excited to hear feedback and thoughts from everyone here on the proposal below!

Abstract

Reflexer Labs has designed a new stable asset called RAI, which is a self-stabilizing, asset-backed credit facility. This proposal aims to introduce RAI to the Tornado community and begin discussion around the potential addition of RAI as a supported asset on Tornado Cash.

After its launch in February, over $100m worth of RAI has been minted and more than $300M worth of ETH has been deposited into the protocol. RAI has managed to stabilize using its on-chain PID controller and is now undergoing a governance minimization process with the intention to harden the protocol.

_ (Fire Eyes DAO) is a collective of Web3 natives navigating token economic and governance systems across the DeFi ecosystem. We’re working alongside the Reflexer Labs team to communicate and coordinate this discussion/integration.

Why RAI?

RAI is taking a new approach at building a trust-minimized, stable asset solely backed by ETH.

Most of today’s stable assets suffer from a level of custodial or centralization risk. The primary example of this is with fiat stablecoins like USDC and USDT where centralized custodians are entirely responsible for holding onto user deposits in their own reserve.

While decentralized alternatives mitigate these risks significantly, alternatives use centralized collteral within their system and/or heavily rely on tokenholder governance (humans!) to maintain the health of the system.

RAI attempts to avoid these pitfalls by moving as much control of the protocol as possible to on-chain, automated algorithms. Backed purely by ETH, RAI is unable to be ‘seized’, switched off or otherwise taken down by any government or regulator. In a similar vein, long-term RAI will have minimal tokenholder governance over crucial system parameters like the redemption rate thanks to the on-chain PID controller.

Benefits to Tornado.Cash

• Ability to transact in a non-sovereign stable asset via tornado.cash in a more censorship resistant and trustless manner.
• Potential for incentivized RAI deposits into Tornado.Cash via FLX ‘ungovernance token’.
• Hardening and increased utility of the Tornado.Cash product due to there being more trustless assets/liquidity available to mix.
• Growth in community and market for Tornado Cash!

TL;DR

• We propose adding RAI to Tornado Cash, using 42,000 RAI AND 420,000 RAI as the denominations.

We believe that RAI is an ideal next step for the Tornado Cash community, given it’s ideological and architectural alignment. Adding RAI pools would give users more flexibility and confidence in depositing into Tornado Cash, given that RAI is resistant to the primary form of attack that current DAI pools could be targeted by.

Next Steps

What do people think? This post was written to spark a conversation and see whether the Tornado Cash community believes that there is value to be created through integrating RAI! Please share your opinion, ideas, concerns and anything else relevant

Hi Callum, welcome to the TC forum.
I am not familiar with RAI. We are very careful with adding new asset pools as bootstrapping them would normally come at the price of decreasing the other pools anonimity set, unless there were pre-existant market demand for a particular stablecoin which would then migrate with it.
I am curious on your point “Potential for incentivized RAI deposits into Tornado.Cash via FLX ‘ungovernance token’.”, could you elaborate on this?

Hey @optionsmate , we tend to focus our incentive campaigns on two things: liquidity and partnerships.

If Tornado would add RAI we’d be happy to collaborate on a strategy to reward RAI depositors with FLX (MKR equivalent for RAI). We usually use off-chain scripts, detect activity and reward every address according to the time & weight of their position.

I am against this honestly. You have to think from a user perspective. Why would you use unknown cryptoassets, when there’s ethereum, wbtc and reputable stablecoins available?

As @Ethermuchacho said, we have several reputable stablecoins available such as DAI, USDC and USDT. Maybe we could use yours when it will have a lot of users.

Hi friends,

I’ve never been formally part of the Tornado Cash team but I’ve had the privilege of contributing at its inception. As summoner of MolochDAO it was my honor to award Tornado with their first ecosystem grant[1] and help fund their trusted setup security audit[2]. It’s given me great joy to see Tornado scale from a few hundred ETH to start approaching flippening ZCash[3] over the last 2 years. As the 2nd ever production zkSNARKS system Tornado.Cash is a systemically important factor in the global debate around individual financial sovereignty. By demonstrating the value of private money[4] and its technological feasibility, Tornado serves as a beacon of hope that Central Bank Digital Currencies (CBDCs) will also employ zkSNARKs to provide comparable anonymity to physical cash. Should they choose not to, however, Tornado serves as a much more profound beacon of hope that non-state sponsored alternative privacy-preserving financial systems will ultimately triumph.

Using Tornado with ETH is fantastic, of course, but as we know stablecoins are growing exponentially in popularity. Most stablecoins, though, don’t share the trust-minimization property that makes ETH so great. USDC/USDT are essentially revocable bank deposits, and DAI is increasingly backed by USDC (close to 50% as of today[5]). As a cofounder of Reflexer, we designed RAI[6] in response to DAI being backed by too much of what we felt was trusted collateral like USDC, in addition to having overreaching governance. RAI is designed to be simple, stable, and predictable. By minimizing the role of governance over RAI, we make it harder for anyone, including ourselves, to mess with the system. We believe that having RAI as an asset in Tornado Cash will serve that niche group of cypherpunks which has multi-year time horizons and wants to build on a reliable financial stack that won’t one day be rugpulled out from under them.

Links (2 max for new users so I broke up the rest):

1. https://dao.molochdao.com/proposals/91
2. https://dao.molochdao.com/proposals/117
3. twitter. com/ameensol/status/1395534483202269187
4. twitter. com/ameensol/status/1343899823523516417
5. daistats. com/#/
6. ethresear. ch/t/announcing-metacoin-the-governance-minimized-decentralized-stablecoin/6897

I highly support this view and the addition of RAI. It’d be a fantastic trust minimized stablecoin option to add. ETH is already the main use case of Tornado currently. Tornado is definitely missing the stablecoin privacy market segment and no matter how much of USDC or DAI it adds, it can’t fully offer the uncensorable guarantees as RAI if predominantly USDC+DAI have regulatory risk. Adding RAI to Tornado and also potentially adding AP rewards for RAI could make Tornado a 1 stop shop for anonymous ETH transactions and anonymous ETH stability transactions through RAI. It would complete Torando’s core offering imo. I’d strongly support this as a founder of another stablecoin project. RAI is the real deal.

I cant agree with this proposal more. DAI is becoming captured by an overly technical bureaucratic structure which is hard to navigate and keep track off. Further, DAI is increasing composed of regulatory captured assets which are exposed to significant risk.

CBDC will surely put people’s privacy under threat IMO as they are structurally ideated to control the population through implementation of MMT and removing the possibility to save, let alone privacy.
Anyway, I agree with DAI’s backing being a matter of concern for some but its popularity made it an asset of choice here. Could you please give an ELI5 on what makes RAI different in relation to less centralized backing?

For the record: in principle I like the idea of adding new assets pool (was also favourable to Sam’s proposal to add FRAX), as fundamentally it widens the breadth of choices to TC users. My only concern is diluting the AS would newly added asset pools not have high enough intrinsic market demand to absorb such dilution.

I support this strongly. I immediately supported reflexer with a vault when i heard RAI would only be backed by eth.
Super important to have this option on ethereum as it looks like we will be getting more and more regulated as the years go by.

Anyone concerned with stabletoken regulations should care to read the documentation and understand, that Reflexer is just as liable to get ran over as Maker or any dollar peged asset is.

Claiming RAI offsets this is absurd.

I think RAI & TORN are a good match given their vision. And the roadmap towards the RAI decentralization could make a better candidate than the other stablecoins.

Yes, RAI is a good project, but I don’t understand why we should propose it in a pool, as there is a little mcap

I support the addition of any asset with provable demand. Tornado should be entirely agnostic of which assets use its code. We don’t care who mixes their money, why should we can what money they mix