OML: A Primitive for Reconciling Open Access with Owner Control in AI Model Distribution

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Cheng, Zerui, Contente, Edoardo, Finch, Ben, Golev, Oleg, Hayase, Jonathan, Miller, Andrew, Moshrefi, Niusha, Nasery, Anshul, Nailwal, Sandeep, Oh, Sewoong, Tyagi, Himanshu, Viswanath, Pramod
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914071987617792
author Cheng, Zerui
Contente, Edoardo
Finch, Ben
Golev, Oleg
Hayase, Jonathan
Miller, Andrew
Moshrefi, Niusha
Nasery, Anshul
Nailwal, Sandeep
Oh, Sewoong
Tyagi, Himanshu
Viswanath, Pramod
author_facet Cheng, Zerui
Contente, Edoardo
Finch, Ben
Golev, Oleg
Hayase, Jonathan
Miller, Andrew
Moshrefi, Niusha
Nasery, Anshul
Nailwal, Sandeep
Oh, Sewoong
Tyagi, Himanshu
Viswanath, Pramod
contents The current paradigm of AI model distribution presents a fundamental dichotomy: models are either closed and API-gated, sacrificing transparency and local execution, or openly distributed, sacrificing monetization and control. We introduce OML(Open-access, Monetizable, and Loyal AI Model Serving), a primitive that enables a new distribution paradigm where models can be freely distributed for local execution while maintaining cryptographically enforced usage authorization. We are the first to introduce and formalize this problem, introducing rigorous security definitions tailored to the unique challenge of white-box model protection: model extraction resistance and permission forgery resistance. We prove fundamental bounds on the achievability of OML properties and characterize the complete design space of potential constructions, from obfuscation-based approaches to cryptographic solutions. To demonstrate practical feasibility, we present OML 1.0, a novel OML construction leveraging AI-native model fingerprinting coupled with crypto-economic enforcement mechanisms. Through extensive theoretical analysis and empirical evaluation, we establish OML as a foundational primitive necessary for sustainable AI ecosystems. This work opens a new research direction at the intersection of cryptography, machine learning, and mechanism design, with critical implications for the future of AI distribution and governance.
format Preprint
id arxiv_https___arxiv_org_abs_2411_03887
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle OML: A Primitive for Reconciling Open Access with Owner Control in AI Model Distribution
Cheng, Zerui
Contente, Edoardo
Finch, Ben
Golev, Oleg
Hayase, Jonathan
Miller, Andrew
Moshrefi, Niusha
Nasery, Anshul
Nailwal, Sandeep
Oh, Sewoong
Tyagi, Himanshu
Viswanath, Pramod
Artificial Intelligence
Cryptography and Security
The current paradigm of AI model distribution presents a fundamental dichotomy: models are either closed and API-gated, sacrificing transparency and local execution, or openly distributed, sacrificing monetization and control. We introduce OML(Open-access, Monetizable, and Loyal AI Model Serving), a primitive that enables a new distribution paradigm where models can be freely distributed for local execution while maintaining cryptographically enforced usage authorization. We are the first to introduce and formalize this problem, introducing rigorous security definitions tailored to the unique challenge of white-box model protection: model extraction resistance and permission forgery resistance. We prove fundamental bounds on the achievability of OML properties and characterize the complete design space of potential constructions, from obfuscation-based approaches to cryptographic solutions. To demonstrate practical feasibility, we present OML 1.0, a novel OML construction leveraging AI-native model fingerprinting coupled with crypto-economic enforcement mechanisms. Through extensive theoretical analysis and empirical evaluation, we establish OML as a foundational primitive necessary for sustainable AI ecosystems. This work opens a new research direction at the intersection of cryptography, machine learning, and mechanism design, with critical implications for the future of AI distribution and governance.
title OML: A Primitive for Reconciling Open Access with Owner Control in AI Model Distribution
topic Artificial Intelligence
Cryptography and Security
url https://arxiv.org/abs/2411.03887