MIT, Stanford, and Princeton:

• Open Letter (350+ co-signers) on the need for AI safe harbors

• A Safe Harbor for AI Evaluation and Red Teaming

• AIR-Bench

• AI Risk Repository

The Hacking Policy Council:

• AI red teaming - Legal clarity and protections needed

• Comments to NIST’s AI Safety Institute on Managing Misuse Risk for Dual-Use Foundation Models

HackerOne:

• An Emerging Playbook for AI Red Teaming With HackerOne

• Anthropic Expands Their Model Safety Bug Bounty Program

BugCrowd:

• Vulnerability Rating Taxonomy

Disclose.io:

• VDP Policy Generator

CMU / Community Emergency Response Team (CERT) / Artificial Intelligence Security Incident Response Team (AISIRT):

• Lessons Learned in Coordinated Disclosure for Artificial Intelligence and Machine Learning Systems

• CERT Guide to Coordinated Vulnerability Disclosure

• On managing vulnerabilities in AI/ML systems

Cybersecurity and Infrastructure Security Agency (CISA): 

• Software Must Be Secure by Design, and Artificial Intelligence Is No Exception

• Secure-by-Design

• CISA and Joint-Seal AI Publications

• Cyber Incident Reporting for Critical Infrastructure Act (CIRCIA) Reporting Requirements

Center for Security and Emerging Technology (CSET): 

• Adversarial Machine Learning and Cybersecurity: Risks, Challenges, and Legal Implications

Data and Society: 

• AI Red-Teaming Is Not a One-Stop Solution to AI Harms: Recommendations for Using Red-Teaming for AI Accountability

Humane Intelligence:

• Generative AI Red Teaming Challenge: Transparency Report

• Algorithmic Risk Assessments, Audits, & Human Rights Report (GNI)

• Red teaming large language models (LLMs) for resilience to scientific disinformation

DeepMind:

• Frontier Safety Framework

• Ethical and social risks of harm from Language Models

• Taxonomy of Risks posed by Language Models

OpenAI:

• Preparedness Framework

• Response to NIST Executive Order on AI

Microsoft:

• How Microsoft Approaches AI Red Teaming

• Microsoft AI Red Team building future of safer AI

Bugcrowd:

• Vulnerability Rating Taxonomy

• Ultimate Guide to Vulnerability Disclosure

OpenPolicy:

• Innovative Leaders' Perspectives on AI Red Teaming

• Comments on National Institute of Standards and Technology (NIST) SP 800-218A: Secure Software Development Practices for Generative AI and Dual-Use Foundation Models

Academic Communities:

• Identifying and Mitigating the Security Risks of Generative AI

• Adversarial Nibbler

DEFCON AI Village (and HuggingFace):

• Coordinated Flaw Disclosure for AI: Beyond Security Vulnerabilities

Hugging Face:

• Red-Teaming Large Language Models

• LLM Evaluation Guidebook

Algorithmic Justice League:

• Change From the Outside: Towards Credible Third-Party Audits of AI Systems

Queer in AI:

• Bound by the Bounty

Knight Columbia First Amendment Institute: 

• A Safe Harbor for Platform Research

Federation of American Scientists:

• A Safe Harbor For AI Researchers: Promoting Safety And Trustworthiness Through Good-Faith Research

The U.S. Department of State:

• Risk Management Profile for Artificial Intelligence and Human Rights

National Institute of Standards and Technology (NIST): 

• NIST AI 600-1- Artificial Intelligence Risk Management Framework: Generative Artificial Intelligence Profile

• NIST Artificial Intelligence Risk Management Framework (AI RMF 1.0)

• NIST AI 800-1 (Initial Public Draft) Managing Misuse Risks of Dual-Use Foundation Models 

• NIST SP 800-218A Secure Software Development Practices for Generative AI and Dual-Use Foundation Models: An SSDF Community Profile

• NIST AI 100-2 E2023: Adversarial Machine Learning: A Taxonomy and Terminology of Attacks and Mitigations

• NIST Cybersecurity Framework (CSF) 2.0

• NIST Vulntology

US Legislative Proposals:

• AI Incident Reporting and Security Enhancement Act

Mozilla:

• External researcher access to closed models

DEFCON AI Village, Generative AI Red Teaming 2:

• To Err is AI : A Case Study Informing LLM Flaw Reporting Practices

MITRE:

• MITRE Adversarial Threat Landscape for Artificial-Intelligence Systems (ATLAS)

• MITRE Common Weakness Enumeration (CWE)

OECD:

• OECD AI Incidents Monitor (AIM)

AIID:

• AI Incident Database (AIID)  

AI Risk and Vulnerability Alliance:

• AI Vulnerability Database (AVID)

Allen Institute for AI/UW:

• WildBench: Benchmarking LLMs with Challenging Tasks from Real Users in the Wild

• WildTeaming at Scale: From In-the-Wild Jailbreaks to (Adversarially) Safer Language Models