What is Sapien?

Sapien is an open infrastructure platform that provides verifiable quality signals for AI systems through blockchain-based consensus mechanisms. It introduces Proof of Quality (PoQ), a trust primitive that enables organizations to prove the credibility and provenance of AI judgments, model outputs, training data, and safety-critical decisions.

The Problem It Solves

As AI systems become increasingly autonomous—from agents making independent decisions to vehicles navigating real-world scenarios—the lack of verifiable quality assurance becomes critical. Every judgment in AI pipelines relies on subjective human evaluation somewhere, whether rating model outputs, labeling training data, or approving safety-critical actions. Currently, there's no way to prove who made these judgments, verify how they were made, or assess source credibility. Sapien fixes this by creating immutable, cryptographically verifiable quality attestations.

Key Features & Capabilities

• Task Definition Specification (TDS): Define what quality means for your specific use case with precise, measurable criteria
• Stake-Based Consensus: Contributors and validators stake tokens and reach economic consensus on quality assessments through commit-reveal mechanisms
• Onchain Attestations: Immutable records of who created content, who validated it, and what consensus was reached—all cryptographically verifiable
• Privacy-Preserving Architecture: Your data never leaves your systems; only attestations go onchain
• Neutral Settlement: Trust that doesn't depend on any single party, backed by blockchain infrastructure

Workflow & Use Cases

Sapien fits into multiple stages of the AI lifecycle:

• Training Data: Verify the quality and labeling of datasets before training models
• Fine-Tuning: Validate preference data at the source with cryptographic proof
• Evaluation: Add verifiable quality gates to evaluation pipelines
• Production: Real-time verification for safety-critical decisions in autonomous systems

The platform operates through a three-step process: (1) Define quality criteria and submit data, (2) Contributors and validators stake and reach consensus, (3) Attestations are written onchain as permanent proof of quality.

Who It's For

Sapien serves AI builders who need verifiable quality assurance, workers who validate and label data, and validators who stake on quality consensus. It's particularly valuable for organizations deploying autonomous agents, safety-critical AI systems, or any application where provenance and trust are paramount.