Eigenlayer has recently released a whitepaper detailing its token, EIGEN. This token is designed to facilitate consensus on data that can't be verified directly on-chain but can be validated through intersubjective agreement. Intersubjective agreement refers to consensus among multiple observers, which may be necessary for applications like prediction markets, storage services, and gaming virtual machines.
For instance, “1 BTC = 1 USD” is a statement that can’t be verified directly on-chain, but can easily be classified as false using off-chain data. The EIGEN token is set to help form consensus around these types of statements.
The concept behind EIGEN builds on previous work by Augur, a platform that introduced the REP token specifically for prediction markets. Unlike REP, EIGEN aims to serve a broader range of applications, positioning itself as a "universal" token for intersubjective agreements. The token is designed to be isolatable from decentralized finance (DeFi) applications during forks, to assist in determining the costs associated with forming social consensus, and to compensate users for losses due to faults.
A key feature of the EIGEN token is its governance mechanism, which supports "intersubjective forking." This enables token holders to address and challenge instances of incorrect slashing.
In the event of a seemingly incorrect slashing, the challenger must stake EIGEN tokens as a bond to initiate a fork. The broader EigenLayer community then votes on whether the fork should proceed. If the community approves the fork, the challenger recovers their bond and receives a reward, while those who voted against the fork may see penalties applied to their tokens. Conversely, if the vote does not favor the fork, the challenger's staked tokens are burned.
EIGEN is set to serve as a component in securing the EigenDA network. Furthermore, the utility of EIGEN may expand if other Actively Validated Services (AVSs) decide to accept EIGEN tokens as staking collateral to secure their networks.
Learn more: What Is EigenLayer?