Risks of ignoring uncertainty propagation in AI-augmented security pipelines
arXiv:2407.14540 · doi:10.1111/risa.70059
Abstract
The use of AI technologies is being integrated into the secure development of software-based systems, with an increasing trend of composing AI-based subsystems (with uncertain levels of performance) into automated pipelines. This presents a fundamental research challenge and seriously threatens safety-critical domains. Despite the existing knowledge about uncertainty in risk analysis, no previous work has estimated the uncertainty of AI-augmented systems given the propagation of errors in the pipeline. We provide the formal underpinnings for capturing uncertainty propagation, develop a simulator to quantify uncertainty, and evaluate the simulation of propagating errors with one case study. We discuss the generalizability of our approach and its limitations and present recommendations for evaluation policies concerning AI systems. Future work includes extending the approach by relaxing the remaining assumptions and by experimenting with a real system.
Accepted for publication in Risk Analysis: An International Journal
References in corpus (9)
- A Review of Uncertainty Quantification in Deep Learning: Techniques, Applications and Challenges
- Aleatoric and Epistemic Uncertainty in Machine Learning: An Introduction to Concepts and Methods
- Do Developers Update Their Library Dependencies? An Empirical Study on the Impact of Security Advisories on Library Migration
- Less Training, More Repairing Please: Revisiting Automated Program Repair via Zero-shot Learning
- Devign: Effective Vulnerability Identification by Learning Comprehensive Program Semantics via Graph Neural Networks
- Automated Patch Assessment for Program Repair at Scale
- DEAR: A Novel Deep Learning-based Approach for Automated Program Repair
- AIBugHunter: A Practical Tool for Predicting, Classifying and Repairing Software Vulnerabilities
- Probability bound analysis: A novel approach for quantifying parameter uncertainty in decision-analytic modeling and cost-effectiveness analysis