From paradigm by Arjun Balaji

Decentralized networks need coordination mechanisms to incentivize and hold their node operators accountable. This began with Proof of Work and later evolved with Proof of Stake, a significant development that allowed networks to source validator security through economic collateral. The next frontier of this is shared security, which expands the services node operators in PoS can provide while making use of the same underlying economic collateral.

Symbiotic is a generalized, permissionless protocol providing shared security through restaking. We are investing in Symbiotic alongside Cyber.Fund, a collaborator of ours on Lido and other protocols.

We believe Symbiotic’s flexible and permissionless approach will be a great fit for many of the most useful consumers of shared security and, over time, could be a default option for bootstrapping a decentralized network.

Background

Lido, Ethereum’s largest liquid staking token, was founded on the insight that separating staking capital from validator infrastructure (labor) is possible without modifying Ethereum’s consensus, using a smart contract layer for routing users’ stake to operators in a decentralized way. This separation, to become “delegated,” is a natural and inherent tendency of Proof of Stake systems. Lido enabled Proof of Stake to scale on Ethereum without compromising decentralization by matching staked Ether with the highest-quality infrastructure operators.

Paradigm began collaborating with Lido protocol contributors in 2021. Since then, Lido has grown from ~$800M to over $36B in staked ETH deposits and fostered the most robust ecosystem of node operators: reputable, geographically distributed, diverse, and aligned.

Paradigm has also long supported the Cosmos ecosystem, leading Tendermint Inc.’s original Series A and later investing in Osmosis and dYdX. Through Cosmos, we observed the challenges of recruiting validators and capital from scratch every time developers wanted to launch a new chain, which significantly limited the pace of innovation.

A natural “phase two” of Ethereum staking is to repurpose stake and validator infrastructure and expertise beyond L1 consensus to secure many protocols at once. This makes it easier to stand up new protocols. Cosmos pioneered this idea as “shared security,” and EigenLayer’s “restaking” recognized that an ETH-centric approach could successfully bootstrap this validator ecosystem. This primitive is novel and powerful, but considering the risks of overloading Ethereum’s consensus, it needs to be designed thoughtfully to enable useful applications safely.

As we considered the market, we realized Konstantin was also interested. Through him, we met Misha and Algys, the founders of Statemind, a top auditor with a close working relationship with Lido (i.e., on their V2 audit), Curve, InstaDapp, and others. We were aligned with how they each viewed the market and jumped at the opportunity to partner.

Introducing Symbiotic

Symbiotic is a new shared security system. It is designed as a thin coordination layer that is maximally flexible, permissionless, and reliable. Symbiotic allows network developers to have complete control over specifying their (re)staking implementation and operator set. Zooming out, the long-term goal of the protocol is to provide primitives that help networks navigate the roadmap to decentralization while prioritizing safety and capital efficiency.

Flexible

Protocols built on Symbiotic can control their collateral assets, rewards, and slashing criteria. Symbiotic will initially focus on staked ETH as the largest pool of stake capital. However, the protocol is general-purpose and can accept any ERC-20 asset as collateral. In time, we expect Symbiotic will service many assets and related operator infrastructure groups.

Symbiotic network developers will also have full control of their operator selection mechanics. Over time, it will be possible to maximize the number of participants, their geographic distribution, and their overlap with other protocols, reputation, and other selection criteria.

Permissionless

The core Symbiotic contracts are immutable, which removes external governance risks. Symbiotic will never have a central multisig, slashing committee, or other permissioning mechanisms for shared security services. Services built on Symbiotic will be able to support many different slashing resolution mechanisms, which we believe is critical for innovation.

Reliable

Building infrastructure operator networks is challenging, as we have learned in working with Lido. Symbiotic will ensure that restaking can scale by onboarding reputable and geographically distributed infrastructure partners and by supporting smaller operators.

What should be built on Symbiotic?

At Paradigm, we view restaking protocols as delegated staking systems at their core. Stakers are incentivized to vote their capital behind operators they believe will be honest validators. In some sense, this is how Ethereum staking (via stETH and other LSTs) already works today.

Near term, we believe that the clearest and safest use case of shared delegated proof of stake security is for bootstrapping new consensus instances:

In the long term, we are also interested in L1 block production use cases, such as new types of MEV auctions, preconfirmations, and based sequencing. However, we think block production use cases may take longer to blossom: they often benefit from adoption by a greater fraction of L1 proposers and may pose more direct security risks to Ethereum L1.

To help achieve that vision, we also created Reth Execution Extensions (ExEx). ExExes allow fast data extraction and processing from a node and enable networks/services to peer with other ExExes to agree on the state that should eventually be injected back into Ethereum. We want to make ExExes the best tool for building shared security services using Symbiotic.

Of course, Symbiotic is a general system, and developers can build any protocol they want on top of it without asking for permission or using our codebases. These are just our own intuitions about the types of use cases most likely to succeed.