
As a part of an ongoing effort to maintain you knowledgeable about our newest work, this weblog put up summarizes some current publications from the SEI within the areas of provide chain threat administration, technical debt, giant language fashions, quantum computing, acquisition, and trustworthiness in AI programs. These publications spotlight the most recent work of SEI technologists in these areas. This put up features a itemizing of every publication, writer(s), and hyperlinks the place they are often accessed on the SEI web site.
The Measurement Challenges in Software program Assurance and Provide Chain Danger Administration
by Nancy R. Mead, Carol Woody, and Scott Hissam
On this paper, the authors talk about the metrics wanted to foretell cybersecurity in open supply software program and the way requirements are wanted to make it simpler to use these metrics within the provide chain. The authors present examples of doubtless helpful metrics and underscore the necessity for knowledge assortment and evaluation to validate the metrics. They assert that defining metrics, amassing and analyzing knowledge for example their utility, and utilizing customary strategies requires unbiased collaborative work to attain the specified outcomes.
Learn the white paper.
Report back to the Congressional Protection Committees on Nationwide Protection Authorization Act (NDAA) for Fiscal 12 months 2022 Part 835 Impartial Research on Technical Debt in Software program-Intensive Programs
by Ipek Ozkaya, Forrest Shull, Julie B. Cohen, and Brigid O’Hearn
A group from SEI performed an unbiased examine to fulfill the necessities of the Fiscal 12 months 2022 Nationwide Protection Authorization Act (NDAA) Part 835, Impartial Research on Technical Debt in Software program-Intensive Programs.
This report describes the conduct of the examine, summarizes the technical tendencies noticed, and presents the ensuing suggestions. The examine methodology features a literature evaluation, a evaluation of SEI studies developed for program stakeholders, deep dives on program knowledge from SEI engagements with Division of Protection (DoD) packages, and interviews performed utilizing the ten examine components laid out in Part 835(b).
The examine concludes that packages are conscious of the significance of managing technical debt. Moreover, a variety of DoD packages have established practices to actively handle technical debt. Throughout this examine, the DoD printed a number of steerage paperwork that start to incorporate technical debt and technical debt administration as a necessary observe for profitable software program improvement. Research suggestions embrace that the DoD should proceed to replace coverage/steerage and empower packages to include technical debt practices as a part of their software program improvement actions whereas enabling analysis in improved device assist and knowledge assortment.
View the report.
Assessing Alternatives for LLMs in Software program Engineering and Acquisition
By Stephany Bellomo, Shen Zhang, James Ivers, Julie B. Cohen, and Ipek Ozkaya
On this white paper, the authors look at how resolution makers, corresponding to technical leads and program managers, can assess the health of enormous language fashions (LLMs) to handle software program engineering and acquisition wants. They introduce exemplar eventualities in software program engineering and software program acquisition, they usually determine frequent archetypes. The authors additionally describe frequent considerations involving the usage of LLMs and enumerate techniques for mitigating these considerations. Utilizing these frequent considerations and techniques, the authors display how resolution makers can assess the health of LLMs for their very own use circumstances by two examples.
Learn the white paper.
The Cybersecurity of Quantum Computing: 6 Areas of Analysis
By Tom Scanlon
Analysis and improvement of quantum computer systems continues to develop at a speedy tempo. The U.S. authorities alone spent greater than $800 million on quantum info science analysis in 2022. Thomas Scanlon, who leads the info science group within the SEI CERT Division, was just lately invited to be a participant within the Workshop on Cybersecurity of Quantum Computing, co-sponsored by the Nationwide Science Basis (NSF) and the White Home Workplace of Science and Know-how Coverage, to look at the rising subject of cybersecurity for quantum computing. On this SEI podcast, Scanlon discusses the right way to create the self-discipline of cyber safety of quantum computing and descriptions six areas of future analysis in quantum cybersecurity.
Hearken to the SEI podcast.
Learn Tom Scanlon’s weblog put up, which offers a technical deep dive into this work.
Connecting Stakeholders for DoD Software program Programs
By Hasan Yasar
The Deputy Secretary of Protection permitted and signed the Division of Protection (DoD) Software program Modernization Technique on February 1, 2022. This act initiated a transformative journey for the division to ship strong software program capabilities on the tempo of evolving calls for. Although system complexity and stakeholder decentralization usually problem our capacity to securely ship capabilities on time and on finances, fostering collaboration and facilitating open discussions about these challenges, strategizing, and devising actionable plans can be essential to overcoming them. This webcast highlights how the December 2023 DoD Weapon Programs Software program Summit options in addressing these challenges and offering us a chance to collectively devise efficient options.
Attendees
- achieve an understanding for software program modernization technique
- uncover difficult areas for DoD weapon programs
- observe and study enablers for fixing system complexity
- align your system’s aims together with your software program engineering practices
- achieve insights from key talks and classes
Measuring the Trustworthiness of AI Programs
By Katherine-Marie Robinson, Carol J. Smith, and Alexandrea Steiner
The flexibility of synthetic intelligence (AI) to companion with the software program engineer, physician, or warfighter relies on whether or not these finish customers belief the AI system to companion successfully with them and ship the end result promised. To construct applicable ranges of belief, expectations should be managed for what AI can realistically ship. On this podcast from the SEI’s AI Division, Carol Smith, a senior analysis scientist specializing in human-machine interplay, joins design researchers Katie Robinson and Alex Steiner, to debate the right way to measure the trustworthiness of an AI system in addition to questions that a company ought to ask earlier than figuring out if it desires to make use of a brand new AI expertise.
Hearken to the podcast.
Learn the associated SEI weblog put up, Contextualizing Finish-Person Wants: Find out how to Measure the Trustworthiness of an AI System, by Carrie Gardner, Katherine-Marie Robinson, Carol J. Smith, and Alexandrea Steiner.