The DoD wants to transform defense acquisitions and enable rapid and iterative delivery of software capability through new acquisition pathways and by exploiting DevSecOps. The emphasis on DevSecOps ...

Additional Sites Directory The following web properties are part of the SEI's work. Some are restricted to approved users, but most are open wikis that offer technical details about ongoing work.

SEI work such as technical reports, logos, articles, presentations, and methods is the intellectual property of Carnegie Mellon University. To request permission to use Carnegie Mellon copyrighted ...

In this “Train, but Verify” project, we will attempt to address the gap in the state of the art on secure training of ML systems with two objectives: Train secure AI systems by training ML models to ...

The DoD is increasingly seeking to deploy AI systems for mission-critical tasks. Modern AI systems most commonly employ machine learning (ML) models to make important, domain-relevant inferences.

The operational complexity of cyber-physical systems (CPS) forces new autonomous features into day-to-day systems, such as vehicles and factories, a phenomenon termed Increasingly Autonomous CPS ...

To address this challenge, we launched this project to develop and implement techniques for automated semantic-equivalence checking. Our tool can identify which decompiled functions are likely to be ...

The SEI has a rich portfolio of research and expertise to add to the national initiative. The following projects provide a view of that portfolio: Train, But Verify – This research in robust and ...

A major gap in achieving quantum advantage is the identification of applications in which quantum computing could provide computational advantage (in terms of time to solution, quality of solution, ...

Dio De Niz - Technical Director, Assuring Cyber-Physical Systems - Technical Director, Assuring Cyber-Physical Systems ...

This project is creating an automated conformance checker that can be integrated into the continuous integration workflow to detect and report nonconformances in hours instead of the months or years ...

Created October 2020 Machine learning (ML) systems have the power and potential to greatly expand our capabilities. Unfortunately, these systems are vulnerable to attacks that seek to exfiltrate ...