This proposal allocates 25% of the Juror Incentive Program (JIP) reward pool to jurors whose staking address is registered on Proof of Humanity (PoH). This bonus portion is distributed proportionally to the square root of the PNK staked by each verified juror. This approach encourages broad human participation while disincentivizing centralization of stake or delegation to a few PoH accounts. The mechanism is additive, sybil-resistant, and active for a 12-month trial period.
Motivation
The JIP has successfully increased PNK staking and juror engagement. This proposal seeks to build on that success by introducing an optional, soft incentive for human-verified participation.
Kleros rightly values pseudonymity. However, encouraging a broader base of uniquely human stakers, without requiring identity or removing pseudonymity, can improve court legitimacy, governance resilience, and sybil resistance. PoH provides a lightweight, opt-in signal for humanity thatâs especially valuable for DAO decision-making and decentralized courts.
This proposal does not change any existing juror behavior, it simply offers a modest boost for those who wish to opt in.
Specification
JIP Pool Split:
25% of the total JIP reward pool per epoch is allocated to a separate PoH Bonus Pool. The remaining 75% is distributed as usual to all stakers per the standard JIP logic.
Eligibility:
Only jurors whose staking address is registered on Proof of Humanity at the start of the epoch are eligible for the PoH Bonus Pool.
Quadratic Distribution Formula:
The PoH Bonus Pool is distributed among PoH-verified addresses proportionally to the square root of their PNK stake. For each eligible juror i:
Trial Duration:
The mechanism will run for a fixed 12-month period. At the end of the trial, the DAO will vote to extend, revise, or sunset it.
Benefits
Improved Sybil Resistance
This mechanism lightly favors human-verified staking, making it more costly for sybil strategies to dominate long-term juror rewards.
Wider Juror Base
Incentivizes staking and participation from a broader set of uniquely human participantsâcomplementing rather than replacing pseudonymous jurors.
Maintains Composability
The program is opt-in, non-restrictive, and does not require changes to existing juror workflows or onboarding processes.
Aligned with Kleros Values
Optionality and decentralization are preserved. This proposal expands participation without reducing access.
Risks and Mitigations
Favoritism Concerns
This is an additive boost. No rewards are removed from existing jurors. Non-PoH users can continue to participate fully.
Doxxing or Identity Exposure
PoH is pseudonymous and does not require government ID or KYC. Stakers voluntarily associate their address with a human profile.
PoH Governance Stability
While PoH has experienced governance challenges, the identity registry remains operational and useful for this limited incentive.
Closing Thoughts
This proposal introduces a modest, low-risk experiment to encourage greater human participation in the Kleros ecosystem, while fully preserving the systemâs permissionless, decentralized, and pseudonymous ethos.
The goal is not to replace pseudonymous participation but to augment Klerosâ robustness by giving uniquely human jurors a small, optional boost.
I think the idea itself is great, even if we should tweak the details.
We must have a way to allow for anonymous participation. Jurors having to reveal their identities can both:
Create an unfair opportunity cost to jurors wishing to have their privacy preserved.
Can make Kleros less resistant toward pressures toward jurors (remember that jurors have been subject to âlegalâ threats in the past, both during 2020 US election dispute and UST depeg disputes).
Anonymous participation can be a zk-proof (linking a PoH ID to another address) or even start simply by trusting the coop not to leak the links.
I think itâs the most interesting proposal weâve seen from non coop members, itâs great to see people getting interested in governance.
This repo has built all the tools to easily integrate a privacy layer over proof of humanity including handling the case where addresses are removed from the registry (see the open issue).
The coop should allocate some developer resources to update the repo to support v2 , especially as essential infrastructure for v2 POH gated courts.
The privacy preserving proof of membership in POH would be required to be renewed periodically, for example every 3 months, to ensure the membership remains valid.
In a ZK-POH gated court, the proof of membership can be produced at the time of voting.
sounds like KIP-79 should be paused and split into two proposals with a hard dependencyâŚ
KIP-ZK (): Should pass first and ship a privacy-preserving PoH attestation (zk-PoH) that lets jurors prove PoH membership without linking staking/voting addresses.
Scope:
Verifier contracts compatible with PoH v2; periodic proof renewal (i.e. every 3 months) baked into the flow.
Court + staking UI integration so proofs are produced at vote time, not account-link time.
Handling for PoH removals/appeals and revocations.
Testnet pilot â audit â mainnet; w/ a coordinator; timeline; rollback plan.
Maintenance of existing implementations as essential infra for PoH-gated courts.
KIP-79 (re-filed after KIP-ZK).
A few suggested tweaks for KIP-79:
Smaller initial split (10â15%) for two epochs, DAO checkpoint before any increase.
Reasoning: Start as a bounded pilot to limit downside and measure real effects. Lower split reduces redistribution risk to non-PoH jurors & makes rollbacks cheap if it underperforms. Also gives the DAO clean data to justify (or reject) scaling later.
KPIs + 12-month sunset tied to juror decentralization and participation; auto-revert if targets miss.
Reasoning: Ties success to explicit security/participation metrics (higher share of unique human jurors, lower stake Gini, stable/improved appeal participation and resolution time). Automatically reverts to baseline JIP if KPIs not met, no extra governance lift needed.
One-human-one-bonus guardrails across addresses; explicit abuse/clawback rules.
Reasoning: Quadratic rewards are vulnerable to address splitting and rotation. Proposal should require per-epoch human uniqueness (across addresses), define how PoH revocations/appeals affect eligibility, add a pause switch for detected abuse, and specify clawbacks to deter gaming.
Reasoning: Remove ambiguity (âmultiplierâ vs âpool splitâ), specify exact parameters and onchain changes, and show payout examples across juror sizes. Deterministic rules improve auditability & transparency, reduce surprises, and empower jurors to model expected outcomes before proposal goes live.
Previous post looks like AI generated so I wonât answer it in details (please if you have some ideas, just put them, no need to make them less clear with undisclosed AI use).
@kleroschad, yeah we can require zk-proofs every 3 months, but only for those wanting anonymity. Users who donât mind can just use the address linked to their PoHID.
It means that if a user is unregistered from PoH, he would still be able to collect for up to 3 months, but canât cumulate addresses (as we link the private address to the PoHID, not the registering address).
The question is:
Do we want to launch that ASAP trusting the coop?
Do we want to wait for a zk solution?
Here the time required for a zk solution would be interesting to know.
Madhur, any take on that?
): Should pass first and ship a privacy-preserving PoH attestation (zk-PoH) that lets jurors prove PoH membership without linking staking/voting addresses.