{
  "@context": {
    "dcterms": "http://purl.org/dc/terms/",
    "dcmitype": "http://purl.org/dc/dcmitype/"
  },
  "dcterms:identifier": [
    "tag:aim-pro.eu,2026:oer/05068b6e469e",
    "https://arxiv.org/abs/2402.06414",
    "arxiv:2402.06414"
  ],
  "dcterms:title": [
    "Trust the Process: Zero-Knowledge Machine Learning to Enhance Trust in Generative AI Interactions"
  ],
  "dcterms:type": [
    {
      "@id": "dcmitype:Text"
    },
    "preprint"
  ],
  "dcterms:creator": [
    "Bianca-Mihaela Ganescu",
    "Jonathan Passerat-Palmbach"
  ],
  "dcterms:description": [
    "Generative AI, exemplified by models like transformers, has opened up new possibilities in various domains but also raised concerns about fairness, transparency and reliability, especially in fields like medicine and law. This paper emphasizes the urgency of ensuring fairness and quality in these domains through generative AI. It explores using cryptographic techniques, particularly Zero-Knowledge Proofs (ZKPs), to address concerns regarding performance fairness and accuracy while protecting model privacy. Applying ZKPs to Machine Learning models, known as ZKML (Zero-Knowledge Machine Learning), enables independent validation of AI-generated content without revealing sensitive model information, promoting transparency and trust. ZKML enhances AI fairness by providing cryptographic audit trails for model predictions and ensuring uniform performance across users. We introduce snarkGPT, a practical ZKML implementation for transformers, to empower users to verify output accuracy and quality while preserving model privacy. We present a series of empirical results studying snarkGPT's scalability and performance to assess the feasibility and challenges of adopting a ZKML-powered approach to capture quality and performance fairness problems in generative AI models."
  ],
  "dcterms:subject": [
    "cs.LG (arxiv)",
    "cs.CR (arxiv)"
  ],
  "dcterms:language": [
    "en"
  ],
  "dcterms:license": [
    {
      "@id": "http://creativecommons.org/licenses/by-sa/4.0/"
    }
  ],
  "dcterms:rights": [
    "CC-BY-SA-4.0 — assessed as SHARE_ALIKE by the harvester's licence gate. Conditions: attribution required, adaptations must carry the same licence. open but ShareAlike/copyleft — derivatives keep the license"
  ],
  "dcterms:accessRights": [
    "open"
  ],
  "dcterms:format": [
    "text/markdown"
  ],
  "dcterms:extent": [
    "42326 bytes (extracted text)"
  ],
  "dcterms:issued": [
    "2024-02-09"
  ],
  "dcterms:provenance": [
    "Retrieved from arXiv on 2026-10-09 in response to the search string “(all:\"artificial intelligence\" OR all:\"machine learning\" OR all:\"generative AI\" OR all:\"deep learning\" OR all:\"reinforcement learning\" OR all:\"large language model\") AND (all:\"AI concepts\" OR all:\"types of AI\" OR all:\"AI fundamentals\" OR all:\"recognizing AI\" OR all:\"recognising AI\" OR all:\"general versus narrow AI\" OR all:\"narrow AI\" OR all:\"general AI\" OR all:\"machine intelligence\" OR all:\"AI strengths and weaknesses\" OR all:\"traditional software\" OR all:\"rule-based systems\" OR all:\"introduction to AI\" OR all:\"introduction to artificial intelligence\" OR all:\"artificial intelligence introduction\" OR all:\"AI primer\" OR all:\"foundations of artificial intelligence\" OR all:\"overview of AI\" OR all:\"understanding AI\" OR all:\"history of AI\" OR all:\"AI essentials\" OR all:\"AI terminology\" OR all:\"metaphors for AI\" OR all:\"AI fundamental concepts\" OR all:\"AI key concepts\" OR all:\"philosophy of AI\" OR all:\"critical AI literacy\")”. arXiv served the resource and is not asserted to be its publisher or author.",
    "Text extracted from latexml-html to Markdown by arxiv-html; the original is retained unchanged beside it."
  ],
  "inferred": {
    "dcterms:references": [
      {
        "value": {
          "@id": "https://arxiv.org/abs/1607.06450",
          "title": "arXiv:1607.06450"
        },
        "method": "inferred",
        "rule": "citation",
        "rule_version": "links/1",
        "evidence": {
          "links": [
            {
              "how": "citation",
              "line": 228,
              "basis": "reference",
              "literal": "arXiv:1607.06450"
            }
          ],
          "kind": "citation"
        }
      },
      {
        "value": {
          "@id": "https://arxiv.org/abs/2109.05662",
          "title": "arXiv:2109.05662"
        },
        "method": "inferred",
        "rule": "citation",
        "rule_version": "links/1",
        "evidence": {
          "links": [
            {
              "how": "citation",
              "line": 235,
              "basis": "reference",
              "literal": "arXiv:2109.05662"
            }
          ],
          "kind": "citation"
        }
      },
      {
        "value": {
          "@id": "https://arxiv.org/abs/2205.14135",
          "title": "arXiv:2205.14135"
        },
        "method": "inferred",
        "rule": "citation",
        "rule_version": "links/1",
        "evidence": {
          "links": [
            {
              "how": "citation",
              "line": 236,
              "basis": "reference",
              "literal": "arXiv:2205.14135"
            }
          ],
          "kind": "citation"
        }
      },
      {
        "value": {
          "@id": "https://arxiv.org/abs/2303.16520",
          "title": "arXiv:2303.16520"
        },
        "method": "inferred",
        "rule": "citation",
        "rule_version": "links/1",
        "evidence": {
          "links": [
            {
              "how": "citation",
              "line": 245,
              "basis": "reference",
              "literal": "arXiv:2303.16520"
            }
          ],
          "kind": "citation"
        }
      },
      {
        "value": {
          "@id": "https://arxiv.org/abs/2303.13375",
          "title": "arXiv:2303.13375"
        },
        "method": "inferred",
        "rule": "citation",
        "rule_version": "links/1",
        "evidence": {
          "links": [
            {
              "how": "citation",
              "line": 258,
              "basis": "reference",
              "literal": "arXiv:2303.13375"
            }
          ],
          "kind": "citation"
        }
      },
      {
        "value": {
          "@id": "https://arxiv.org/abs/2303.08774",
          "title": "arXiv:2303.08774"
        },
        "method": "inferred",
        "rule": "citation",
        "rule_version": "links/1",
        "evidence": {
          "links": [
            {
              "how": "citation",
              "line": 259,
              "basis": "reference",
              "literal": "arXiv:2303.08774"
            }
          ],
          "kind": "citation"
        }
      },
      {
        "value": {
          "@id": "https://arxiv.org/abs/2205.11584",
          "title": "arXiv:2205.11584"
        },
        "method": "inferred",
        "rule": "citation",
        "rule_version": "links/1",
        "evidence": {
          "links": [
            {
              "how": "citation",
              "line": 260,
              "basis": "reference",
              "literal": "arXiv:2205.11584"
            }
          ],
          "kind": "citation"
        }
      },
      {
        "value": {
          "@id": "https://arxiv.org/abs/1906.07221",
          "title": "arXiv:1906.07221"
        },
        "method": "inferred",
        "rule": "citation",
        "rule_version": "links/1",
        "evidence": {
          "links": [
            {
              "how": "citation",
              "line": 261,
              "basis": "reference",
              "literal": "arXiv:1906.07221"
            }
          ],
          "kind": "citation"
        }
      }
    ]
  }
}