Accès ouvert déclaré
2026
preprint
Muse Spark Safety & Preparedness Report
Cristina Menghini, Peter Ney, Hamza Kwisaba, Zifan, Wang, Miles Turpin, Felix Binder, Jean-Christophe Testud, Aidan Boyd, Nathaniel Li, Ivan Evtimov, Klaudia Krawiecka, Arman Zharmagambetov, Jeremy Kritz, Alexander R. Fabbri, Daniel Song, Jinpeng Miao, Joonas Hjelt, Meghna Ramani, Leona Lan, Reza Aghajani, Joanna Bitton, Mahesh Pasupuleti, Devin Norder, Khalid El-Arini, Paridhi Singh, Vítor Albiero, Sahana CB, Rashnil Chaturvedi, Elahe Dabir, Edoardo Debenedetti, Jim Gust, Ziwen Han, Kat He, Sean Hendryx, Lifeng Jin, Polina Kirichenko, Sandra Lefdal, Kenneth Li, Asad Liaqat, Inna Lin, Despoina Magka, Neal Mangaokar, Ishita Mediratta, Zach Miller, Smitha Milli, Niloofar Mireshghallah, Saba Nazir, Hung Nguyen†, Maximilian Nickel, Kelvin Niu, Kerem Oktar, Bhargavi Paranjape, Parth Pathak, Maya Pavlova, Emmanuel Ramirez, David Renardy, Candace Ross, Yasha Sheynin, Claudia Shi, Shivam Singhal, Evangelia Spiliopoulou, Rakshith Sharma Srinivasa, Jamelle Watson-Daniels, Spencer Whitman, Adina Williams, Chen Xing, Andy Zou, Ma Tc, Siqi Deng, James Beldock, Prashant Ratanchandani, Kate Plawiak, Taesung Lee, Ryan Victory, Lindsay Hundley, Rachad Alao, Himaghna Bhattacharjee, Jianfeng Chi, Gary Frost, Pegah Ghahremani, Niki Howe, Yuheng Huang, Saeed Jahed, Hannah Korevaar, Trang Le, Z LIU, Jun Luo, Qin Lyu, Nina Mehrabi, Abraham Montilla, Chirag Nagpal, Cyrus Nikolaidis, Rajvardhan Oak, Manoj Ravi, Vidya Sarma, Aman Shankar, Alana Shine, Eric Michael Smith, Mariana Tandon, Michael Tontchev, 王曹宇, Zihan Wang, Corinne Wong, Zheng Wu, Hongyuan Zhan, Justin Zhao, Zexuan Zhong, Chengxu Zhuang, Tristan Goodman, Ayaz Minhas, Harrison Rudolph, Victoria Jeffries, Ingrid Dickinson, Alex Vaughan, Lauren Deason, Kamalika Chaudhuri, Julian Michael, Shengjia Zhao, Summer Yue
0Citations signalées, ce qui n’est pas une note de qualité
0Institutions déclarées
0Pays d’affiliation déclarés
Le résumé fourni par la source
Muse Spark is the latest large language model developed by Meta. In this report, we first present evaluations for catastrophic risk domains under Meta's Advanced AI Scaling Framework, along with the evidence that informed our launch decision. We then discuss additional considerations, such as Muse Spark's broader content safety and behavioral profile, that are relevant to overall safety but fall outside the catastrophic risk domains governed by the Framework. Our preparedness results covering Chemical and Biological, Cybersecurity, and Loss of Control risks assess Muse Spark's deployment within Meta AI as presenting acceptable levels of residual risks under our Advanced AI Scaling Framework. We conducted a broad set of evaluations targeting dual-use and high-risk capabilities across these catastrophic risk domains. Those evaluations identified elevated risks prior to mitigations, with Chemical and Biological capabilities assessed as likely reaching the "high risk" category under the Advanced AI Scaling Framework before safeguards were applied. We have implemented a multi-layered set of mitigations that address the identified risks, and Muse Spark demonstrates state-of-the-art refusal across a range of benchmarks related to hazardous workflows in chemistry and biology. We therefore release Muse Spark as the underlying model of Meta AI.
Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.
Le contrôle bibliographique ouvert
Aucun DOI disponible pour le contrôle Crossref.
Les sujets associés
Adversarial Robustness in Machine LearningScientific Computing and Data ManagementSafety Systems Engineering in Autonomy