Vassilev, A.
(2025),
Robust AI Security and Alignment: A Sisyphean Endeavor?, IEEE Security & Privacy, [online], https://doi.org/10.1109/MSEC.2026.3678214, https://tsapps.nist.gov/publication/get_pdf.cfm?pub_id=960858 (Accessed May 28, 2026)
Official websites use .gov
A .gov website belongs to an official government organization in the United States.
Secure .gov websites use HTTPS
A lock (
) or https:// means you’ve safely connected to the .gov website. Share sensitive information only on official, secure websites.
Robust AI Security and Alignment: A Sisyphean Endeavor?
Published
Author(s)
Abstract
This article establishes information-theoretic limitations for robustness of artificial intelligence (AI) security and alignment. Knowing these limitations and preparing for their challenges is essential for responsible adoption of AI. Broader implications for cognitive reasoning limitations of AI systems are also proven.Citation
IEEE Security & Privacy
Volume
24
Issue
3
Pub Type
Journals
Download Paper
Keywords
Artificial Intelligence, Weapons, Nuclear Weapons, Filtering, Filters, Circuits And Systems, Contacts, Radio Access Networks, Regional Area Networks, Videos, Video Equipment, Less Than Or Equal, Proof Of Theorem, Set Of Rules, Robust System, Finite Set, Broader Implications, Good And Bad, Intelligence Agencies, Binary String, Result Of Theorem, Human Language, Artificial Intelligence Systems, String Length, Cognitive Reasoning, Real Life Systems, Context Window, Adversarial Robustness, Information Theoretic Security, Chatbot
Citation
Issues
If you have any questions about this publication or are having problems accessing it, please contact [email protected].
Created December 12, 2025, Updated May 14, 2026