Artificial Intelligence (AI) is advancing rapidly, presenting complex challenges for ensuring system security. Although AI systems are being widely adopted across sectors, comprehensive frameworks for safeguarding these technologies remain underdeveloped. To address this gap and characterize the emerging discipline of AI security, the National Academies of Sciences, Engineering, and Medicine is conducting a workshop under the auspices of its Forum on Cyber Resilience. The workshop and resulting issue paper bring together experts from across the research community to systematically assess current and emerging threats to AI systems, as well as the risks introduced by their deployment and application. This workshop objectively examines both current and emerging risks to AI systems and the risks that arise from their use. By analyzing notable successes and failures, workshop participants identify and characterize key AI security attributes, making their relevance clearer and more tangible to the public. The resulting issue paper developed from the workshop outlines a forward-looking AI security research agenda for the community. This foundational reference will help coordinate and guide efforts toward meaningful progress in addressing AI security challenges. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.