Skip to content

The CCS Interview

Haven1 director: DeFi's $8B in losses is blocking institutions

Akash Mahendra is a director at the Haven1 Foundation, which is building an EVM-compatible layer one with security guardrails and identity checks built into the network. In this October 2023 conversation with Ashton Addison, he covers institutional hesitancy, how much of the last DeFi cycle was speculation, and the project's testnet plans. At the time of recording, he said the team hoped to launch a public incentivized testnet within about a month.

Interview by Ashton Addison Recorded October 2023 6 min read

Key takeaways

  • Haven1 is an EVM-compatible Layer 1 blockchain implementing network-level guard rails including contract whitelisting and governance approval to prevent hacks and exploits.
  • Over eight billion dollars has been lost to hacks and exploits in DeFi over recent years, with TVL in DeFi protocols near three-year lows despite growth in stablecoins.
  • Haven1 requires identity verification for users and builders before on-chain transactions, storing only verification results on-chain to ensure compliance and known actors.
  • Institutions express interest in DeFi but face concerns about exploit risk, liquidity risk, and counterparty risk, leading many to build private enterprise blockchains instead.
  • Current DeFi landscape remains convoluted with daily vulnerabilities affecting retail users, regulatory uncertainty, and lack of widespread adoption despite blockchain's evolution.

Why DeFi needs built-in guardrails

Ashton Addison: What is Haven1, and what have you built so far?▶ 00:24

Akash Mahendra: Haven1 is an EVM-compatible layer one blockchain. Its goal is to provide secure, accessible, compliant, user-friendly on-chain finance, not only for retail users but also for developers and institutional investors. Despite its evolution over the years, DeFi remains a convoluted and turbulent landscape. Every day, users face hacks, exploits, liquidity risks and rug pulls. There's also added regulatory uncertainty from America, parts of Europe and parts of Asia. That has caused a lack of widespread faith in blockchain technology, and it hasn't delivered the consistent adoption we all hope for. We want to expand the current financial ecosystem and drive more retail and institutional adoption of Web3, through our security frameworks and network-level guardrails. The aim is to enable on-chain finance for every single person.

Ashton: How specifically can Haven1 make DeFi safer?▶ 02:09

Akash: Security is baked into the protocol itself. Our network-level guardrails exist to protect, not restrict, the financial opportunity of using the network. For example, contract deployment is controlled on-chain. Any contract that needs to be deployed requires whitelisting and approval by the validators, and it has to pass through a governance process. Our plan is that every protocol and contract must pass an extensive risk model that we developed, meet industry standards and have certain levels of audits. That should automatically prevent a significant number of the hacks and exploits happening in DeFi today. Even today, the average user doesn't understand the risks and can't read Solidity contracts, so we need to take on the responsibility of securing the network and creating standards for protocols to adhere to, rather than continuing with the Wild West of the past few years. I think more than $8 billion has been lost in DeFi and Web3 in the past couple of years, based on on-chain analysis data, and that number keeps increasing year on year. I don't see how we get widespread adoption, or institutions coming in, until we fix that.

we need to take on the responsibility of securing the network and creating standards for protocols to adhere to

Institutions, risk and identity

Ashton: Where do institutions currently stand on DeFi?▶ 05:24

Akash: There's definitely a lot of interest. During my time on the Yield App DeFi team, we had discussions with institutions such as JPMorgan, BlackRock and Goldman Sachs as they explored blockchain. All of them expressed interest in deploying on-chain, but they had concerns about liquidity risk, the risk of exploits and counterparty risk. I recently saw a figure showing that TVL in DeFi protocols is near a three-year low. That's shocking given the advances of the past three years, while on-chain assets like stablecoins, tokenized gold, silver and stocks keep growing. Personally, I've deployed less of my own capital in DeFi because there's too much uncertainty. Even a protocol everyone assumed was virtually risk-free turned out to have a zero-day exploit in the Vyper compiler, and Balancer, one of the biggest and most trustworthy protocols, has had two or three issues this year. That's why a lot of institutions use, or are building, their own enterprise blockchains. Ideally we need a mixture of public and private chains, with risk controls on a public blockchain. That's where I think we need to go as an ecosystem.

Ashton: What did the Tornado Cash ruling mean for institutions, and how does Haven1 respond?▶ 09:48

Akash: As I understand it, OFAC said that anyone who used that application, or even mixed money with people who did, was non-compliant and might face regulatory issues. Imagine you're an institution like JPMorgan with standards to uphold. You can't deploy money into a public DeFi pool, because you don't know who you're mixing money with. That's too much counterparty risk. You need to know that everyone you interact with on-chain is a known party and compliant with AML requirements and terrorist financing rules. I just don't think institutions can ever deploy to a public blockchain such as Ethereum. I love Ethereum, and I love the financial freedom and inclusivity it gives us, but the bigger players can't do what I can do. So we require all users, builders and anyone deploying capital on-chain to complete identity verification before executing transactions. Only the result of the verification and the country IDs are stored on-chain, and none of the rest of the data is stored by us. Because every user has a proof-of-ID NFT, developers can restrict or enable users from specific regions with a single line of code.

Speculation versus lasting value

Ashton: How much of the last DeFi cycle was speculation rather than sustainable value?▶ 13:38

Akash: I feel 99% of the previous cycle was fueled by speculation. The alternative layer ones, the Ethereum competitors, did extremely well because of the high-gas-fee narrative and Ethereum's limited transactions per second, back when L2s hadn't scaled properly. Then there was the entire NFT narrative and virtual worlds like Decentraland and The Sandbox. It doesn't make sense for a virtual world protocol to be worth $10 billion. I'm pretty sure the bigger ones reached $10 billion-plus fully diluted valuations with barely any users, and they still don't have many. The alternative layer ones generated far fewer fees than Ethereum does today, even in a bear market, so there was no reason to consider them competitors, even though they took TVL from Ethereum for a while. They pumped because they had really good incentive programs. Now we're back to reality, and asset prices have dropped significantly.

Ashton: Which on-chain use cases will outlast the speculation?▶ 17:44

Akash: What is very clear is that there are so many useful things we can do on-chain. Trading on-chain makes everything more transparent and brings more people in. I live in Australia, and if I want to trade tokenized American stocks, I can't do it easily. I have to apply to different brokers, and there may be restrictions placed on me. Ideally, a protocol brings tokenized stocks on-chain, to Ethereum or to a more permissioned, secure platform such as Haven1, and I can do it in a couple of clicks. Even if I have to verify my identity, I should be able to access that as easily as if I lived in the US, or more easily, because I don't need an intermediary. Another example is sportsbooks: a transparent one where you're not getting wrecked by the spread of the odds, and which can be restricted to certain parts of the world if needed. I believe the real value will come out in the next run. But the dumbest things usually end up on top, so I'm not holding my breath. I'm happy that a lot of what's being built now is more useful than the next fork that has no value and is just a fad.

the dumbest things usually end up on top, so I'm not holding my breath.

Testnet, GoQuorum and getting involved

Ashton: Where does the Haven1 network stand today?▶ 20:37

Akash: It's been just over a year since we started architecting and building Haven1. We deployed the private testnet early this year. With it, we designed how the network permissions and parameters would work, the role validators would play, and the best way to implement the identity framework at network level. Since then, we've scaled up the team and onboarded a significant number of partners and developers. We now have an internal public testnet that we're stress testing while we work out the kinks. We're hopeful we can release the public incentivized testnet, a significant milestone, in the next month or so. No promises, though, because this hasn't been done before. Haven1 is built using GoQuorum, an open-source blockchain platform originally built by JPMorgan and currently managed by Consensus. We work closely with the Consensus team every week to upgrade and improve the client. Only through GoQuorum can we create a blockchain with flexible and expressive network permission management. That can't be done on any other L2 or on Ethereum. We do expect some issues to come up in testing, which is all part of the game. Once the public testnet is out, we'll work hard to onboard developers. There will be a grants program, and we plan to run our own hackathon before the end of the year.

Ashton: Will non-developers and DeFi enthusiasts be able to explore the incentivized testnet?▶ 23:55

Akash: Definitely. Once we launch the fully public testnet, there will be campaigns people can join, as well as our own internal program that incentivizes use of the testnet. We need to create the entire flow, so a lot of work has gone into the user experience. I'm always keen to jump on a new testnet or mainnet launch, but it's a confusing process even for me. When Base launched about three or four weeks ago, I was asking what was available and whether there was a lending market. And that's from Coinbase, probably the biggest company in the space. We want a smooth, interactive, easy experience aimed at onboarding current users and people who have never used on-chain finance. If I struggle to find information about a new blockchain, I can't imagine how newer users feel. They'd probably fall for a lot of scams.

This interview has been edited for length and clarity. Recorded October 2023.

Akash Mahendra
Akash MahendraAkash Mahendra is Director of the Haven1 Foundation, which supports Haven1, an EVM-compatible layer 1 blockchain built for secure on-chain finance. He leads strategy, operations and risk management for the project. He started his career as a Legal Enforcement Officer at the Australian Securities and… Full profile →
Share this interviewXLinkedIn
▶ Watching the interview