Yogev Bar-On

As frontier artificial intelligence (AI) models — that is, models that match or exceed the capabilities of the most advanced models at the time of their development — become more capable, protecting them from theft and misuse becomes more critical. Especially important are a model's weights — the learnable parameters derived by training the model on massive data sets. Stealing a model's weights gives attackers the ability to exploit the model for their own use.

Funding for this research provided by gifts from RAND supporters. The research was conducted by the Meselson Center within RAND Global and Emerging Risks. This report is part of the RAND research report series. RAND reports present research findings and objective analysis that address the challenges facing the public and private sectors. All RAND reports undergo rigorous peer review to ensure high standards for research quality and objectivity.

Some time ago, we noticed at Realmode Labs that Amazon Kindle has an interesting feature called “Send to Kindle”. This feature allows Kindle users to send e-books to their device as email attachments. We immediately thought of the potential security concerns of this feature: what if we can send malicious e-books to unsuspecting users? Istarted researching the potential issues, and sure enough, found three vulnerabilities which together allow an attacker to remotely execute code on the Kindle.