Organization · Company
Attestable
A company developing zero-knowledge proofs for AI systems; it reports a prover for large language model inference and proposes proof-based compute accounting.
Attestable describes itself as "The AI Integrity Company" and says it provides zero-knowledge proofs for AI systems. It has published the following on verification:
- Inference prover. Attestable reports a zero-knowledge prover for production-scale transformer inference whose security rests only on cryptographic hash functions 1. See Attestable zero-knowledge inference prover and zero-knowledge proofs of inference.
- Proof-based accounting. It proposes combining inference proofs with proof-of-work accounting, and notes that the verifier "needs a credible estimate of the compute available" to the actor 2; see Proofs of useful work and resource exhaustion.
- Verification firewall. It proposes a firewall at the external interfaces of an environment that holds model weights, which would select outputs for proof at random after each output is committed 3.
Implementations
- Attestable's zero-knowledge prover, which the company reports proves large language model outputs came from committed weights at tens of tokens per second.
Mechanisms
Mechanisms this organization has designed, built, evaluated or supplied.
- A prover produces a cryptographic proof that an output came from running a committed model on a given input, without revealing the weights.
- Cryptographic evidence that hardware performed a given amount of agreed computation, proposed as a way to show no spare capacity remained for other work.
Publications
Sources this organization authored or published.
- CAttestable (2026). From Verifiability to Model-Weight Security. Attestable blog. RecordCited by Zero-knowledge proofs of inference; Attestable zero-knowledge inference prover; Attestable
- CAttestable (2026). Pacing AI Requires Proof. Attestable blog. RecordCited by Proofs of useful work and resource exhaustion; Zero-knowledge proofs of inference; Attestable zero-knowledge inference prover; Pearl proof-of-useful-work blockchain; Declared hardware is idle or shut down; There is no undeclared relevant compute; Proof of (useful) work; Attestable
- CAttestable (2026). Proving LLMs at Scale. Attestable blog. RecordCited by Zero-knowledge proofs of inference; Attestable zero-knowledge inference prover; Attestable
Sources
- CAttestable (2026). Proving LLMs at Scale. Attestable blog. Source recordSupports: zero-knowledge prover for production-scale transformer inference; hash-only security (provider-reported)
- CAttestable (2026). Pacing AI Requires Proof. Attestable blog. Source recordSupports: proposal to combine inference proofs with proof-of-work accounting; need for a compute estimate
- CAttestable (2026). From Verifiability to Model-Weight Security. Attestable blog. Source recordSupports: verification-firewall proposal with randomly selected proofs