_version_ 1866916661690368000
author Longpre, Shayne
Klyman, Kevin
Appel, Ruth E.
Kapoor, Sayash
Bommasani, Rishi
Sahar, Michelle
McGregor, Sean
Ghosh, Avijit
Blili-Hamelin, Borhane
Butters, Nathan
Nelson, Alondra
Elazari, Amit
Sellars, Andrew
Ellis, Casey John
Sherrets, Dane
Song, Dawn
Geiger, Harley
Cohen, Ilona
McIlvenny, Lauren
Srikumar, Madhulika
Jaycox, Mark M.
Anderljung, Markus
Johnson, Nadine Farid
Carlini, Nicholas
Miailhe, Nicolas
Marda, Nik
Henderson, Peter
Portnoff, Rebecca S.
Weiss, Rebecca
Westerhoff, Victoria
Jernite, Yacine
Chowdhury, Rumman
Liang, Percy
Narayanan, Arvind
author_facet Longpre, Shayne
Klyman, Kevin
Appel, Ruth E.
Kapoor, Sayash
Bommasani, Rishi
Sahar, Michelle
McGregor, Sean
Ghosh, Avijit
Blili-Hamelin, Borhane
Butters, Nathan
Nelson, Alondra
Elazari, Amit
Sellars, Andrew
Ellis, Casey John
Sherrets, Dane
Song, Dawn
Geiger, Harley
Cohen, Ilona
McIlvenny, Lauren
Srikumar, Madhulika
Jaycox, Mark M.
Anderljung, Markus
Johnson, Nadine Farid
Carlini, Nicholas
Miailhe, Nicolas
Marda, Nik
Henderson, Peter
Portnoff, Rebecca S.
Weiss, Rebecca
Westerhoff, Victoria
Jernite, Yacine
Chowdhury, Rumman
Liang, Percy
Narayanan, Arvind
contents The widespread deployment of general-purpose AI (GPAI) systems introduces significant new risks. Yet the infrastructure, practices, and norms for reporting flaws in GPAI systems remain seriously underdeveloped, lagging far behind more established fields like software security. Based on a collaboration between experts from the fields of software security, machine learning, law, social science, and policy, we identify key gaps in the evaluation and reporting of flaws in GPAI systems. We call for three interventions to advance system safety. First, we propose using standardized AI flaw reports and rules of engagement for researchers in order to ease the process of submitting, reproducing, and triaging flaws in GPAI systems. Second, we propose GPAI system providers adopt broadly-scoped flaw disclosure programs, borrowing from bug bounties, with legal safe harbors to protect researchers. Third, we advocate for the development of improved infrastructure to coordinate distribution of flaw reports across the many stakeholders who may be impacted. These interventions are increasingly urgent, as evidenced by the prevalence of jailbreaks and other flaws that can transfer across different providers' GPAI systems. By promoting robust reporting and coordination in the AI ecosystem, these proposals could significantly improve the safety, security, and accountability of GPAI systems.
format Preprint
id arxiv_https___arxiv_org_abs_2503_16861
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle In-House Evaluation Is Not Enough: Towards Robust Third-Party Flaw Disclosure for General-Purpose AI
Longpre, Shayne
Klyman, Kevin
Appel, Ruth E.
Kapoor, Sayash
Bommasani, Rishi
Sahar, Michelle
McGregor, Sean
Ghosh, Avijit
Blili-Hamelin, Borhane
Butters, Nathan
Nelson, Alondra
Elazari, Amit
Sellars, Andrew
Ellis, Casey John
Sherrets, Dane
Song, Dawn
Geiger, Harley
Cohen, Ilona
McIlvenny, Lauren
Srikumar, Madhulika
Jaycox, Mark M.
Anderljung, Markus
Johnson, Nadine Farid
Carlini, Nicholas
Miailhe, Nicolas
Marda, Nik
Henderson, Peter
Portnoff, Rebecca S.
Weiss, Rebecca
Westerhoff, Victoria
Jernite, Yacine
Chowdhury, Rumman
Liang, Percy
Narayanan, Arvind
Artificial Intelligence
The widespread deployment of general-purpose AI (GPAI) systems introduces significant new risks. Yet the infrastructure, practices, and norms for reporting flaws in GPAI systems remain seriously underdeveloped, lagging far behind more established fields like software security. Based on a collaboration between experts from the fields of software security, machine learning, law, social science, and policy, we identify key gaps in the evaluation and reporting of flaws in GPAI systems. We call for three interventions to advance system safety. First, we propose using standardized AI flaw reports and rules of engagement for researchers in order to ease the process of submitting, reproducing, and triaging flaws in GPAI systems. Second, we propose GPAI system providers adopt broadly-scoped flaw disclosure programs, borrowing from bug bounties, with legal safe harbors to protect researchers. Third, we advocate for the development of improved infrastructure to coordinate distribution of flaw reports across the many stakeholders who may be impacted. These interventions are increasingly urgent, as evidenced by the prevalence of jailbreaks and other flaws that can transfer across different providers' GPAI systems. By promoting robust reporting and coordination in the AI ecosystem, these proposals could significantly improve the safety, security, and accountability of GPAI systems.
title In-House Evaluation Is Not Enough: Towards Robust Third-Party Flaw Disclosure for General-Purpose AI
topic Artificial Intelligence
url https://arxiv.org/abs/2503.16861