Imagen de Portada

Expanding External Access To Frontier AI Models For Dangerous Capability Evaluations

Fuente: arXiv

Guardado en:

Detalles Bibliográficos
Autores principales:	Charnock, Jacob, Tlaie, Alejandro, O'Brien, Kyle, Casper, Stephen, Homewood, Aidan
Formato:	Preprint
Publicado:	2026
Materias:	Computers and Society
Acceso en línea:	Acceder al recurso
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

Ejemplares similares

Securing External Deeper-than-black-box GPAI Evaluations
por: Tlaie, Alejandro, et al.
Publicado: (2025)

Using AI Alignment Theory to understand the potential pitfalls of regulatory frameworks
por: Tlaie, Alejandro
Publicado: (2024)

What Should Frontier AI Developers Disclose About Internal Deployments?
por: Charnock, Jacob, et al.
Publicado: (2026)

Assessing confidence in frontier AI safety cases
por: Barrett, Stephen, et al.
Publicado: (2025)

Audit Cards: Contextualizing AI Evaluations
por: Staufer, Leon, et al.
Publicado: (2025)

The AI Risk Spectrum: From Dangerous Capabilities to Existential Threats
por: Grey, Markov, et al.
Publicado: (2025)

A Blueprint for an EU Ecosystem of Secure, Deep and External AI Audits
por: Tlaie, Alejandro
Publicado: (2025)

A Methodology for Quantitative AI Risk Modeling
por: Murray, Malcolm, et al.
Publicado: (2025)

Coordinated Disclosure of Dual-Use Capabilities: An Early Warning System for Advanced AI
por: O'Brien, Joe, et al.
Publicado: (2024)

Frontier Lag: A Bibliometric Audit of Capability Misrepresentation in Academic AI Evaluation
por: Gringras, David, et al.
Publicado: (2026)

The Role of Risk Modeling in Advanced AI Risk Management
por: Touzet, Chloé, et al.
Publicado: (2025)

Asymmetry by Design: Boosting Cyber Defenders with Differential Access to AI
por: Ee, Shaun, et al.
Publicado: (2025)

Pitfalls of Evidence-Based AI Policy
por: Casper, Stephen, et al.
Publicado: (2025)

Frontier AI Auditing: Toward Rigorous Third-Party Assessment of Safety and Security Practices at Leading AI Companies
por: Brundage, Miles, et al.
Publicado: (2026)

Risk Reporting for Developers' Internal AI Model Use
por: Delaney, Oscar, et al.
Publicado: (2026)

Randomness, Not Representation: The Unreliability of Evaluating Cultural Alignment in LLMs
por: Khan, Ariba, et al.
Publicado: (2025)

Practical Principles for AI Cost and Compute Accounting
por: Casper, Stephen, et al.
Publicado: (2025)

Exploring and steering the moral compass of Large Language Models
por: Tlaie, Alejandro
Publicado: (2024)

Evaluating AI Providers' Frontier Safety Frameworks
por: Stelling, Lily, et al.
Publicado: (2025)

Third-party compliance reviews for frontier AI safety frameworks
por: Homewood, Aidan, et al.
Publicado: (2025)

Technical Requirements for Halting Dangerous AI Activities
por: Barnett, Peter, et al.
Publicado: (2025)

Evaluating Frontier Models for Dangerous Capabilities
por: Phuong, Mary, et al.
Publicado: (2024)

Manipulation and the AI Act: Large Language Model Chatbots and the Danger of Mirrors
por: Krook, Joshua
Publicado: (2025)

Frontier AI Ethics: Anticipating and Evaluating the Societal Impacts of Language Model Agents
por: Lazar, Seth
Publicado: (2024)

'Teens Need to Be Educated on the Danger': Digital Access, Online Risks, and Safety Practices Among Nigerian Adolescents
por: Oguine, Munachimso B., et al.
Publicado: (2025)

The Safety Gap Toolkit: Evaluating Hidden Dangers of Open-Source Models
por: Dombrowski, Ann-Kathrin, et al.
Publicado: (2025)

The California Report on Frontier AI Policy
por: Bommasani, Rishi, et al.
Publicado: (2025)

The Evaluation Differential: When Frontier AI Models Recognise They Are Being Tested
por: Vishwarupe, Varad, et al.
Publicado: (2026)

Sabotage Evaluations for Frontier Models
por: Benton, Joe, et al.
Publicado: (2024)

Toward Quantitative Modeling of Cybersecurity Risks Due to AI Misuse
por: Barrett, Steve, et al.
Publicado: (2025)

Assurance of Frontier AI Built for National Security
por: Pistillo, Matteo, et al.
Publicado: (2025)

Catastrophic Liability: Managing Systemic Risks in Frontier AI Development
por: Kierans, Aidan, et al.
Publicado: (2025)

Limits of Safe AI Deployment: Differentiating Oversight and Control
por: Manheim, David, et al.
Publicado: (2025)

Expert Survey: AI Reliability & Security Research Priorities
por: O'Brien, Joe, et al.
Publicado: (2025)

Frontier AI developers need an internal audit function
por: Schuett, Jonas
Publicado: (2023)

Trends in Frontier AI Model Count: A Forecast to 2028
por: Kumar, Iyngkarran, et al.
Publicado: (2025)

The Role of AI Safety Institutes in Contributing to International Standards for Frontier AI Safety
por: Fort, Kristina
Publicado: (2024)

Open Problems in Frontier AI Risk Management
por: Ziosi, Marta, et al.
Publicado: (2026)

Towards Safe Multilingual Frontier AI
por: Kanepajs, Artūrs, et al.
Publicado: (2024)

Responsible Reporting for Frontier AI Development
por: Kolt, Noam, et al.
Publicado: (2024)