analiza aiqbench
This paper introduces TP-CRIV, a novel framework for verifying AI model identity in black-box scenarios where the verifier lacks direct access to the claimant's model. By utilizing fresh, undisclosed challenges under network isolation, the method provides statistical evidence that a specific model is locally possessed without relying on cryptographic proofs or provider cooperation. The approach addresses critical concerns regarding model misappropriation and unauthorized deployment of remote AI services.
TP-CRIV: A Framework for Third-Party Challenge-Response Identity Verification of AI Models
arXiv:2609.29264v1 Announce Type: cross Abstract: Artificial intelligence (AI) models are increasingly deployed through remote services, making model misappropriation a growing concern. Existing approaches, including watermarking, fingerprinting, and model similarity analysis, primarily rely on predefined evidence or direct behavioral comparison and do not explicitly evaluate whether the claimant currently possesses and can utilize model-dependent information relevant to the claimed model identity. In this paper, we propose Third-Party Challenge-Response Identity Verification (TP-CRIV) for AI models. TP-CRIV targets a third-party verification setting in which the verifier has neither white-box nor API access to the claimant's model, can interact with the suspicious deployed service only through its ordinary black-box inference interface, and does not require protocol-specific cooperation from the service provider. Under these constraints, the framework enables the verifier to obtain empirical evidence as to whether the claimant locally possesses a model satisfying a predeclared identity relative to the deployed model. Verification is conducted under fresh, previously undisclosed requirements and network isolation, so that the demonstrated capability cannot rely on online external assistance after challenge disclosure. The resulting evidence is interpreted relative to independently specified and calibrated matching and non-matching operating situations and is statistical rather than cryptographic. We instantiate TP-CRIV for CNN image classifiers using probability-control-based witness generation. Experiments on ten ImageNet-pretrained TorchVision models demonstrate clear same/cross-model separation and finite-challenge verification using independently calibrated thresholds.
Abstract:Artificial intelligence (AI) models are increasingly deployed through remote services, making model misappropriation a growing concern. Existing approaches, including watermarking, fingerprinting, and model similarity analysis, primarily rely on predefined evidence or direct behavioral comparison and do not explicitly evaluate whether the claimant currently possesses and can utilize model-dependent information relevant to the claimed model identity. In this paper, we propose Third-Party Challenge-Response Identity Verification (TP-CRIV) for AI models. TP-CRIV targets a third-party verification setting in which the verifier has neither white-box nor API access to the claimant's model, can interact with the suspicious deployed service only through its ordinary black-box inference interface, and does not require protocol-specific cooperation from the service provider. Under these constraints, the framework enables the verifier to obtain empirical evidence as to whether the claimant locally possesses a model satisfying a predeclared identity relative to the deployed model. Verification is conducted under fresh, previously undisclosed requirements and network isolation, so that the demonstrated capability cannot rely on online external assistance after challenge disclosure. The resulting evidence is interpreted relative to independently specified and calibrated matching and non-matching operating situations and is statistical rather than cryptographic. We instantiate TP-CRIV for CNN image classifiers using probability-control-based witness generation. Experiments on ten ImageNet-pretrained TorchVision models demonstrate clear same/cross-model separation and finite-challenge verification using independently calibrated thresholds.
Cryptography and Security (cs.CR); Artificial Intelligence (cs.AI)
arXiv:2609.29264 [cs.CR]
(or arXiv:2609.29264v1 [cs.CR] for this version)
https://doi.org/10.48550/arXiv.2609.29264 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Text integral sub licența deschisă a publicației · 315 cuvinte
Subiecte conexe
Characterizing Statistical Separability in TP-CRIV for Probabilistic AI Models
arXiv · 10 oct. 2026
Cloud Forensics and the Digital Crime Scene: How Investigation, Prevention, and Resilience Have Evolved
AppDirect · 21 mai 2026
Social Media Bans Aren’t Enough to Make Children Safe
IEEE Spectrum · 25 sept. 2026
Natera’s intelligent appointment scheduling with Amazon Bedrock AgentCore
AWS · 26 aug. 2026