mlsanityData to 14 Oct 2026

AI Security Research Should Better Incentivize Defense Research

arXiv 2605.23448 ↗First posted 22 May 2026cs.CR cs.AIcited by 0

Not on the map. Why?

Abstract (arXiv)

This work examines an imbalance in artificial intelligence (AI) security research: the field tends to produce more work on attacking AI systems than on defending them. Drawing on related academic papers, we find biased attack-to-defense ratios across subfields, including federated learning, speech recognition, membership inference, large language models, etc. The imbalance possibly means far beyond a simple count: attack papers are routinely evaluated under favorable conditions that make threats look more severe than they are in practice, while defenses are held to a stricter standard that few can meet. The result is a literature rich in demonstrated vulnerabilities and thin on usable and deployed protections. We thus argue that AI security research should better incentivize defense research.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 29 of its 42 references that are in the map. First posted on arXiv 22 May 2026.

Builds on

29 of 42 references.

References · 29

  1. 2013Intriguing properties of neural networkscited by 5,155
  2. 2017One pixel attack for fooling deep neural networkscited by 582
  3. 2021Membership Inference Attacks on Machine Learning: A Surveycited by 321
  4. 2022A Survey of Machine Unlearningcited by 312
  5. 2024Jailbreak Attacks and Defenses Against Large Language Models: A Surveycited by 230
  6. 2020SoK: Certified Robustness for Deep Neural Networkscited by 105
  7. 2024Safeguarding Large Language Models: A Surveycited by 78
  8. 2022I Know What You Trained Last Summer: A Survey on Stealing Machine Learning Models and Defencescited by 75
  9. 2022Physical Adversarial Attack meets Computer Vision: A Decade Surveycited by 67
  10. 2024Deepfake Generation and Detection: A Benchmark and Surveycited by 67
  11. 2024SoK: Membership Inference Attacks on LLMs are Rushing Nowhere (and How to Fix It)cited by 59
  12. 2022SoK: Let the Privacy Games Begin! A Unified Treatment of Data Inference Privacy in Machine Learningcited by 51
17 more
  1. 2023Exploring the Landscape of Machine Unlearning: A Comprehensive Survey and Taxonomycited by 51
  2. 2020SoK: The Faults in our ASRs: An Overview of Attacks against Automatic Speech Recognition and Speaker Identification Systemscited by 46
  3. 2024Threats, Attacks, and Defenses in Machine Unlearning: A Surveycited by 39
  4. 2024Membership Inference Attacks and Defenses in Federated Learning: A Surveycited by 31
  5. 2024A Survey on Speech Deepfake Detectioncited by 27
  6. 2025SoK: Evaluating Jailbreak Guardrails for Large Language Modelscited by 21
  7. 2023Physical Adversarial Attacks For Camera-based Smart Systems: Current Trends, Categorization, Applications, Research Challenges, and Future Outlookcited by 20
  8. 2025Membership Inference Attacks on Large-Scale Models: A Surveycited by 18
  9. 2021SoK: Anti-Facial Recognition Technologycited by 16
  10. 2023Physical Adversarial Attacks for Surveillance: A Surveycited by 15
  11. 2024The Federation Strikes Back: A Survey of Federated Learning Privacy Attacks, Defenses, Applications, and Policy Landscapecited by 13
  12. 2024SoK: On Gradient Leakage in Federated Learningcited by 12
  13. 2022Visual Adversarial Attacks and Defenses in the Physical World: A Surveycited by 11
  14. 2023SoK: Pitfalls in Evaluating Black-Box Attackscited by 6
  15. 2023SoK: Unintended Interactions among Machine Learning Defenses and Riskscited by 6
  16. 2024SoK: A Review of Differentially Private Linear Models For High-Dimensional Datacited by 5
  17. 2025SoK: Critical Evaluation of Quantum Machine Learning for Adversarial Robustnesscited by 3

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2026To Defend Against Cyber Attacks, We Must Teach AI Agents to Hack
  2. 2026The Attack and Defense Landscape of Agentic AI: A Comprehensive Survey
  3. 2025Cybersecurity AI: Evaluating Agentic Cybersecurity in Attack/Defense CTFs
  4. 2025Offensive Security for AI Systems: Concepts, Practices, and Applications
  5. 2026Cyber-Capable AI Agents: Vulnerabilities, Evaluation Containment, and Defensive Response
  6. 2026The Autonomy Tax: Defense Training Breaks LLM Agents
  7. 2024The AI risk repository: A meta-review, database, and taxonomy of risks from artificial intelligence
  8. 2025The Automation Advantage in AI Red Teaming