3. Consensus Algorithm

3.1. In-Depth Analysis of PRODAO's PoC (Proof of Contribution) Consensus Algorithm

The PRODAO Mainnet secures both the security and efficiency of the network through its proprietary consensus algorithm, PoC (Proof of Contribution). PoC compensates for the energy inefficiency of the existing Proof of Work (PoW) method and the capital-concentration problem of the Proof of Stake (PoS) method, and provides a mechanism that directly rewards “contribution,” a core value of the PRODAO ecosystem. Rather than favoring nodes that simply hold many assets, it aims to induce nodes that create real value for and actively participate in the ecosystem to gain block-generation authority, thereby building a healthier and fairer network.

3.1.1. Contribution Score Criteria and Calculation Logic

The core of the PoC algorithm depends on how fairly and accurately each node's “Contribution Score (CS)” is calculated. The CS is calculated by quantitatively and qualitatively evaluating the various activities an expert performs within the PRODAO ecosystem.

Key Metrics and Weights for Measuring Contribution

The CS is calculated based on the following four major metrics, each of which is assigned a weight according to its importance to the PRODAO ecosystem.

Amount of DID Information Added and Qualitative Evaluation (Weight: 40%)
  • Measurement method: Evaluates the amount of information an expert adds to their DID (Decentralized Identifier) (e.g., field of expertise, career, certifications, degrees, etc.) and the qualitative reliability of that information (e.g., third-party attestation, blockchain-based verifiability).
  • Significance: Because a DID is an expert's core profile and the basis of trust, accurate and rich DID information is regarded as the most important contribution to raising the trustworthiness of the expert ecosystem.
Content Interaction (Weight: 30%)
  • Measurement method: Measures the engagement of other users (views, likes, comments, shares, citations, etc.) with content an expert creates or shares on the platform (posts, papers, projects, etc.).
  • Significance: The creation of high-quality content and active interaction are important contributions that raise the informational value of the ecosystem and promote communication among users.
Data Sharing (Weight: 20%)
  • Measurement method: Measures the amount and utilization of data that an expert, with their own consent, shares within the ecosystem in anonymized or de-identified form.
  • Significance: Shared data becomes a core resource that is used for AI training, the development of new services, and more, increasing the overall value of the ecosystem.
Frequency and Influence of Community Participation (Remaining Weight)
  • Measurement method: Evaluates active participation and positive influence within the community, such as forum participation, governance voting, bug reporting, and ecosystem promotion activities.
  • Significance: Community activity is an essential element for the sustainable growth and development of the ecosystem.

Final Contribution Score (CS) Calculation

The final CS is calculated by multiplying the measured value for each metric by its weight and summing the results. The CS is updated periodically and is designed to reflect the expert's continuous contributions. If a specific malicious act is detected, the CS may be reduced.

Transparency and Fairness

The CS calculation logic is transparently disclosed within the smart contract so that all participants can verify it. This secures the fairness of contribution evaluation and provides clear guidelines on how participants can raise their contributions.

3.1.2. PoC-Based Block Generation and Verification Process

The PoC consensus algorithm maintains the stability of the network by generating and verifying blocks through the following steps.

Signer Node Selection

The PRODAO Mainnet periodically selects, in accordance with the PoC consensus algorithm, the top nodes with the highest Contribution Score (CS) in the network into the Signer node pool. This process combines voting and randomness elements to secure fairness and unpredictability.

Transaction Collection and Block Proposal

Among the selected Signer nodes, the node currently holding block-generation authority (the leader node) collects the valid transactions propagated across the network and proposes a new block containing them. The proposed block includes the hash value of the previous block, the list of transactions, a timestamp, and the signature of the proposing Signer node.

Block Verification and Consensus

The proposed block is propagated to the other Signer nodes. Each Signer node independently verifies the validity of the proposed block (transaction validity, signature validity, etc.). When a certain number or more of the Signer nodes (e.g., two-thirds or more) confirm the block's validity and sign it, consensus is reached.

Block Finalization and Chain Addition

The block for which consensus has been completed is finally added to the blockchain and finalized. A finalized block cannot be reversed, and it is synchronized to all Full Nodes to reflect the latest state of the network.

Reward Payout and Contribution Update

The Signer node that successfully generated the block and the Signer nodes that participated in the consensus receive PROD tokens as a reward. At the same time, the Contribution Score (CS) of all nodes that contributed to block generation and network maintenance is updated. This provides participating nodes with a continuous incentive to contribute.

3.1.3. Prevention of Malicious Behavior and Security Mechanisms (Slashing, etc.)

The PoC consensus algorithm includes powerful mechanisms that prevent malicious behavior in order to maintain the security and integrity of the network.

Staking

A node wishing to participate as a Signer node must stake a certain amount of PROD tokens as collateral. This imposes an economic constraint—that the staked tokens may be forfeited if the node behaves maliciously—thereby inducing honest participation.

Slashing

  • Definition: A powerful penalty mechanism that forfeits some or all of the staked PROD tokens if a Signer node commits a malicious act that harms the network—such as double-signing or improper block proposal/verification—or violates the rules of the consensus network protocol.
  • Objective: To provide a strong deterrent against malicious behavior and to maintain the security and reliability of the network. Slashed tokens may be burned or returned to the network reward pool.
  • How it works: The targets of slashing are validators (PRODAO's Signer nodes) that have deposited a stake as collateral in order to participate in the consensus process. These nodes stake PROD tokens to participate in the network, and it is these staked PROD tokens that are subject to slashing.
  • Trigger conditions: Slashing is triggered when a specific violation specified in the network's consensus rules (e.g., inclusion of an invalid transaction, an invalid signature) is detected. Such violations are usually detected automatically by smart contracts or the network protocol.
  • Penalty: Depending on the severity and type of violation, a certain proportion of the staked tokens (e.g., 1%–100%) is forfeited.
  • In PRODAO's case, some of the forfeited (slashed) PROD tokens are burned, contributing to reducing the total circulating supply and increasing scarcity, while the remainder may be redistributed elsewhere in the ecosystem (e.g., the community treasury, the reward pool).
  • Procedure: Detection of the violation and submission of evidence; approval of slashing by network consensus; slashed tokens are burned (inducing deflation) or moved to the network reward pool.
  • Reward rules: Incentives that validators receive when they contribute to the healthy participation of the network.
  • Block-proposal reward: Paid to the validator that successfully proposed a block.
  • Block-verification reward: Paid to validators that voted on a valid block and contributed to consensus.
  • Reward form: Paid in PROD tokens.
  • Reward distribution: Per block generation, at the end of each epoch, and so on.

Contribution Score (CS)-Based Penalties

In addition to slashing, if behavior that negatively affects network operation (e.g., frequent offline status, attempts to propose defective blocks, etc.) is detected, the Contribution Score (CS) of the relevant node may be reduced. A CS reduction imposes a disadvantage on obtaining future block-generation authority, inducing continuous good-faith participation. Through these multi-layered security mechanisms, PRODAO's PoC consensus algorithm achieves both strong security and efficiency.

3.1.4. Implementation Method for the Participant Contribution and Reputation System Algorithm

This is an algorithm that is essential for implementing the core logic to build a participant contribution and reputation system on the PRODAO Mainnet, and for maintaining the health of the ecosystem by evaluating participants' activities, providing rewards according to contribution, and sanctioning malicious behavior.

In short, this system performs the following functions.

  • Identity verification and eligibility check: Confirms whether a participant is eligible to contribute to the PRODAO Mainnet.
  • AI-based abuse detection: AI performs a first-pass filter to check whether there is anything suspicious in a contribution activity.
  • Contribution-score calculation: Considering the participant's trustworthiness (based on reputation and VC), calculates and accumulates a score for the actual contribution.
  • Application of abuse penalties: When abuse is confirmed, imposes penalties such as contribution-score reduction, staked-token slashing, and participation restrictions.
  • Abuse-severity determination: Judges the severity of abuse based on reported data.

Definitions of Key Symbols and Variables

  • P: A specific participant.
  • Ai: Contribution activity type i.
  • Di: The detailed data of contribution activity i.
  • CS(P): The current total Contribution Score of participant P.
  • Rep(P): The current reputation score of participant P (range: [0,1]).
  • VC(P): The VC (Verifiable Credential)-based trust level of participant P (range: [0,1]).
  • SPROD(P): The quantity of PROD tokens staked by participant P.
  • CAbuse: The severity of confirmed abuse (e.g., 'LOW', 'MEDIUM', 'HIGH', 'CRITICAL').

Main Algorithm Functions and Formulas

Contribution-Activity Processing and Contribution-Score (CS) Calculation Function: fProcessContribution(P, Ai, Di)

This function defines the process of handling participant P's contribution activity Ai and data Di and updating the Contribution Score (CS).

  • Identity and eligibility verification (DID/VC-based): VDID(P) in {0,1}: validity of participant P's DID (0 = invalid, 1 = valid). VVC(P) in {0,1}: whether participant P holds a valid VC (0 = not held, 1 = held). Condition: unless VDID(P) = 1 and VVC(P) = 1 (or the minimum requirements are met), the contribution is rejected.
  • AI-based initial abuse detection: fAI(Di) in [0,1]: an AI-based quality/abuse-suspicion score for the contribution data Di (the closer to 1, the more normal). Condition: if fAI(Di) < TAI_Suspicion, that contribution enters a REPORT_FOR_REVIEW state and awaits further review (TAI_Suspicion is the AI suspicion threshold).
  • Base activity score calculation: Sbase(Ai, Di). A base score is calculated according to the type of each contribution activity Ai and the quantitative/qualitative metrics of the data Di. Example: Sbase(data provision, Ddata) = α1 × (data volume) + α2 × (data quality).
  • Trust coefficient calculation: CTrust(P). Calculated by integrating participant P's reputation score Rep(P) and VC-based trust level VC(P). CTrust(P) = β1 × Rep(P) + β2 × VC(P) (where β1, β2 are DAO-defined weights). CTrust(P) in [0,1].
  • Final activity contribution score calculation: Sfinal(P, Ai, Di) = Sbase(Ai, Di) × CTrust(P). The final score is calculated by multiplying the base activity score by the participant's trust coefficient.
  • Total Contribution Score (CS) update: Each contribution type Ai is assigned a weight W(Ai) defined by DAO governance. CSnew(P) = CSold(P) + Sfinal(P, Ai, Di) × W(Ai). CS(P) is updated on the blockchain.
Abuse-Penalty Application Function: fApplyPenalty(P, CAbuse)

This function applies penalties to a participant P for whom abuse has been confirmed, according to the severity CAbuse.

  • CS (Contribution Score) reduction: CSreduced(P) = max(0, CS(P) − ΔCS(CAbuse)). ΔCS(CAbuse): the amount of CS reduction according to severity (e.g., k1 × CS(P) for 'LOW', k2 × CS(P) for 'HIGH'). k1 < k2 < ... < kN (DAO-defined ratios).
  • Slashing of staked PROD tokens: ΔSPROD(CAbuse): the quantity of PROD tokens to be slashed according to severity. SPROD_slashed(P) = SPROD(P) × SlashingRate(CAbuse). SlashingRate(CAbuse): 1.0 (100%) for 'CRITICAL', 0.5 (50%) for 'HIGH', etc. PerformSlashing(P, SPROD_slashed(P)): executes the forfeiture of tokens on the blockchain.
  • Participation restriction: BanType(CAbuse): the type of participation restriction according to severity (e.g., 'NONE', 'TEMPORARY', 'PERMANENT'). SetParticipationBan(P, BanType(CAbuse)): sets the participation-restriction state.
  • Reputation-score update: Repnew(P) = max(0, Rep(P) − ΔRep(CAbuse)). ΔRep(CAbuse): the amount of reputation-score reduction according to severity.
Abuse-Detection and Severity-Determination Function: fDetermineAbuseSeverity(P, Dreported)

This function determines whether abuse has occurred and its severity by synthesizing AI-analysis results, community reports, cross-verification data, and the like. (This process proceeds through a combination of AI models, decentralized oracles, DAO governance voting, etc.)

  • Input: P (participant), Dreported (reported activity data).
  • Output: CAbuse in {'NONE', 'LOW', 'MEDIUM', 'HIGH', 'CRITICAL'}.
  • Blockchain integration: The participants map should in practice be managed as a state variable of a blockchain smart contract, and data updates should be made through transactions.
  • Oracle and AI integration: Functions such as AI_Model.predictQualityScore, AI_Model.analyzeAbuse, DAO_Governance.getVoteResult, and DecentralizedOracle.verify require API integration with actual AI services, decentralized oracle networks, and the DAO governance system.
  • Weights and thresholds: All weights and thresholds—α1, α2, β1, β2, W(Ai), TAI_Suspicion, k1, k2, ..., SlashingRate, ΔRep, and so on—are designed to be flexibly adjustable through DAO governance.