Zero-Knowledge Proofs

Follow this topic Ask about this topic

Zero-knowledge proofs let a prover convince a verifier that a computation or statement is correct while keeping inputs or other private details hidden. Common approaches include zkSNARKs and Bulletproofs for compact proofs, GKR protocols and zkVMs for computation, and commitments that bind claims to data. Applications include privacy-preserving aggregation, model and training verification, fairness audits, and provenance checks, with benchmarks measuring proof size, proving time, and verification cost. Structured computations are becoming more practical to verify, but costs remain high, claims depend on trustworthy data sources and explicit assumptions, and general oracle-aided computation has fundamental limits.

Latest papers

13 papers1 paper in the last four weeks, with none the four weeks before. 0.0% of all new papers.

Can AI Oversight Be Zero Knowledge?
Oct 1, 2026Alessandro Chiesa, Ziyi Guan, Burcu YildizScalable Oversight Zero-Knowledge Proofs
EPFL · MIT
Zero-Knowledge Predicate Proofs Between AI Agents: A Measured, Cross-Protocol Gateway and the Source-Integrity Gap
Aug 30, 2026Ashok Subbabhatta GopalakrishnaAgent Communication Protocols MCP Security
independent researcher
Certified but Private: Scalable Zero-Knowledge Proofs for Neural Network Guarantees
Aug 17, 2026Youwei Zhong, Ben Merbaum, Timos Antonopoulos +4Certified Robustness Neural Network Verification
Yale University · University of Illinois Urbana-Champaign
Soft Redaction of Image Provenance via Zero-Knowledge Proofs
Aug 7, 2026Muhammad Awan, John CollomosseData Provenance Zero-Knowledge Proofs
University of Surrey, Guildford GU2 7XH, UK · Adobe Research, San Jose CA 95110, USA
ZK-SR117: A Chunked Zero-Knowledge Attestation Design for Aggregated Fair-Lending Metrics, with a Control Mapping toward Full SR 11-7 Coverage
Aug 1, 2026Mohammad Nasir Uddin, Rahnuma Tabassum Orpita, Eklachur Rahman Bhuiyan +1AI Risk Management Algorithmic Auditing
1DBA-BIDA (Business Intelligence and Data Analytics) Program, Westcliff University, USA · Department of Computer Science and Engineering, Northern University Bangladesh, Bangladesh · 3Washington University of Science and Technology, Alexandria, Virginia, USA +1
Agree on the Model, Verify the Inference: GKR Protocols for HND-Based Transformer Inference
Jul 23, 2026Xiaolong Liang, Juanjuan Li, Rui Qin +1Neural Network Verification Transformer Inference
State Key Laboratory of Multimodal Artificial Intelligence Systems, Institute of Automation, Chinese Academy of Sciences, Beijing 100190, China
Proposals and challenges for zero-knowledge verification of frontier AI training
Jun 3, 2026Pierre Peigné, Ky Nguyen, Paul WangZero-Knowledge ML AI Governance
General-Purpose AI Policy Lab Paris · Sorbonne Université, CNRS, LIP6, F-75005 Paris, France
SS-ZKR: Spatial-Semantic Zero-Knowledge Routing for Privacy-Preserving Multi-Agent Collaboration
May 31, 2026Hassan TouheedAgent Communication Protocols Multi-Agent System Security
Privacy-Preserving Federated Learning: Integrating Zero-Knowledge Proofs in Scalable Distributed Architectures
May 4, 2026Divya GuptaByzantine-Robust Federated Learning Privacy-Preserving ML
Independent Researcher Hyderabad, India
Enabling AI ASICs for Zero Knowledge Proof
Apr 20, 2026Jianming Tong, Jingtian Dang, Simon Langowski +6Zero-Knowledge Proofs
Georgia Institute of Technology · Massachusetts Institute of Technology · Google
Fine-Tuning Integrity for Modern Neural Networks: Structured Drift Proofs via Norm, Rank, and Sparsity Certificates
Apr 6, 2026Zhenhang Shang, Yingzhe Yu, Kani ChenNeural Network Verification LLM Auditing
The Hong Kong University of Science and Technology, Hong Kong, China
TAPAS: Efficient Two-Server Asymmetric Private Aggregation Beyond Prio(+)
Mar 20, 2026Harish Karthikeyan, Antigoni PolychroniadouSecure Aggregation Secure Multi-Party Computation
JPMorgan AI Research, JPMorgan AlgoCRYPT CoE
Zero-Knowledge Federated Learning with Lattice-Based Hybrid Encryption for Quantum-Resilient Medical AI
Mar 3, 2026Edouard LansiauxQuantum Federated Learning Privacy-Preserving Federated Learning
Department of Emergency Medicine & STaR-AI Research Group, CHU de Lille, Université de Lille, Lille, France

Oct 1, 2026J/K move · Enter open · S save

Can AI Oversight Be Zero Knowledge?

Alessandro Chiesa, Ziyi Guan, Burcu Yildiz

EPFL · MIT

Scalable Oversight Zero-Knowledge Proofs Verifiable Outsourced Computation

Open paper AskPDF arXiv

AI systems increasingly produce outputs from confidential data, such as a fitness-for-duty assessment from medical records or the predicted properties of a drug candidate from its secret structure. It is important to verify that such outputs are correct without revealing the underlying data. A recent line of work studies verification of AI outputs via interactive proofs and debate for oracle-aided computation, where correctness may depend on an oracle such as human judgment, a physical experiment, or the web. These works focus on verification by a verifier that runs much faster than the computation. However, such efficient verification is impossible for general oracle-aided computation, and these works therefore rely on additional assumptions. We focus instead on privacy: allowing the verifier to run in time polynomial in the computation, we ask whether interactive arguments for oracle-aided computation can be zero knowledge, so that the verifier learns nothing about the confidential data beyond the correctness of the output. We prove that, in general, they cannot. In the random oracle model, there are no zero-knowledge proofs for all oracle-aided computations, even if both the prover and the verifier are allowed to run much longer than the computation itself. The impossibility extends to debate, a canonical model for scalable oversight. On the positive side, we show that if the oracle attaches a cryptographic signature to each of its answers, then every oracle-aided computation can be verified in zero knowledge with an efficient prover and verifier, assuming only collision-resistant hash functions. Beyond privacy, this also gives an alternative approach to scalable oversight that relies neither on an honest opponent, as in debate, nor on the robustness of the computation, as in prior single-prover protocols.