Lagrange
Zero-knowledge proof infrastructure for verifiable AI compute · Lagrange
Lagrange builds cryptographic proof systems (DeepProve) that generate verifiable proofs for every AI inference, allowing mathematical verification without exposing models or data. Like Attestable, they use zero-knowledge proof technology adapted from blockchain to create trust infrastructure for AI systems. Their focus on proving AI computation integrity through cryptographic methods directly parallels Attestable's wedge, though they emphasize compute verification over multi-agent interaction provenance.
Why: Nearly identical technical approach using ZK proofs for AI verification. Main difference: Lagrange focuses on compute/inference verification while Attestable emphasizes multi-agent provenance and data tampering prevention. Both are creating pre-standard trust infrastructure.
Origin Matters
Decision provenance infrastructure with cryptographic verification · Origin Matters
Origin Matters provides cryptographic verification of how AI decisions are formed, enabling proof without exposing sensitive data—directly aligned with Attestable's provenance and privacy-preserving verification approach. They target the same compliance/governance buyer and emphasize decision auditability. While their marketing emphasizes 'decision provenance' over 'multi-agent systems,' the core wedge of cryptographic verification for AI integrity with data protection is nearly identical.
Why: Very close match on cryptographic verification for AI provenance without data exposure. Both target compliance officers and use crypto primitives for trust. Difference: Origin Matters emphasizes decision-level provenance while Attestable focuses on multi-agent interactions, but the technical and market positioning overlap significantly.
Lemma
Trust infrastructure API for AI provenance and cryptographic proofs · Lemma
Lemma provides cryptographic proof infrastructure for AI provenance, decisions, authority, and regulatory attributes through an API. They explicitly position as 'trust infrastructure for AI' using cryptographic proofs, closely matching Attestable's wedge. Their focus on provenance proofs and regulatory compliance aligns with the same buyer persona and problem timing, though their broader scope (authentication, authority) suggests a wider middleware layer than Attestable's agent-specific focus.
Why: Strong overlap on cryptographic trust infrastructure for AI with provenance focus. Both target regulatory compliance. Difference: Lemma offers broader trust primitives (authentication, authority) beyond just AI integrity verification, suggesting a wider but potentially less focused wedge than Attestable's agent-specific approach.
Model Witness
Cryptographic audit infrastructure for AI inference · Model Witness
Model Witness provides tamper-evident logging for AI inference pipelines using ECDSA signing and Merkle-anchored records on blockchain (Polygon). They create cryptographic proof of what models said and when, targeting the same audit/compliance buyer persona. While they use blockchain-derived cryptography like Attestable, their approach emphasizes tamper-evident logs and signatures rather than zero-knowledge proofs, making it a related but distinct technical bet.
Why: Similar cryptographic verification wedge and audit-focused buyer persona. Key difference: uses Merkle trees and signatures rather than ZK proofs, and focuses on inference logging rather than multi-agent provenance. Both target enterprise compliance officers.
