At a time when technological advances are rapidly reshaping the national security landscape, the Department of Defense (DOD) and the Intelligence Community (IC) are leading efforts to harness the power of trusted artificial intelligence (AI). Leidos recently had the opportunity to sit down at the ACT-IAC Emerging Technologies Conference for an informal discussion with two of the leading minds in the field.
Andy MacDonald, Leidos’ CTO for Decision Advantage Solutions, moderated a discussion with Dr. William Streilein, CTO of the DOD’s Chief Digital and AI Office (CDAO), and Dr. John Beieler, AI Director for the Office of the Director of National Intelligence. The experts shared their guiding principles and reflections on the efforts to date. Some key takeaways from the discussion transcript are below.
Bring data to where analytics are done
THE Reference architecture of the CDAO data mesh provides a blueprint to guide and constrain instantiations of data mesh solution architectures, providing department stakeholders with a common language and validation against proven reference architectures (RAs). In doing so, CDAO develops connection points between stakeholder RAs while adhering to a common set of models that bring data to where analytics are built, enabling diverse dashboards and decision support. Moving forward, CDAO’s goal is to make these datasets accessible to the entire department in a self-service manner to accelerate and scale the outcomes of decision benefits toward DOD’s digital transformation goals.
Investing in the entire stack
The IC has been engaged in an AI journey for decades, and investments in this area must strengthen its ability to provide relevant information to policymakers in a timely manner. By understanding what algorithms can be built today, the IC is reimagining what the overall architecture of analyst workflows should look like in the future. Another goal of the IC is to invest in making its technology stack more effective and efficient, enabling better task execution, collection, processing, exploitation, and dissemination.
Related Reading: Artificial Intelligence Ethics Principles for the Intelligence Community
United efforts to ensure AI
Although they operate under different authorities, DOD and IC collaborate in areas such as AI assurance and data infrastructure to support the warfighter. While their approaches differ, important commonalities remain. National security organizations share best practices using a common lexicon so stakeholders know how they can leverage those best practices.
Maturity model for human-machine teamwork
Currently, the Extended Language Model (LLM) features are useful for back-office tasks and legacy code updates that have low consequences. As users move toward higher-consequence use cases, a maturity model will help stakeholders understand what needs to be done to protect workflows. CDAO is developing a Human-Machine Collaboration Maturity Model, as a rubric, with five levels so that operators understand the limitations associated with the model and how they might be successfully exploited.
Related Readings: Trusted AI: The Leidos Method
Commitment to AI Ethical Principles
Adversaries seek to exploit AI, and they may not be bound by the same legal standards, civil liberties, and privacy guidelines as the United States. However, the United States’ commitment to the ethical and responsible use of AI is what makes it a leader in the AI space. Existing published AI principles include ensuring that AI is fair and traceable. Aligning DOD and IC on AI ethical guardrails allows them to innovate quickly and effectively to ensure ethical AI.
Related Reading: Pentagon Official Lays Out DOD’s Vision for AI
New paradigm in industrial collaboration
CDAO’s AI strategy builds on lessons learned from industry, particularly in the area of machine learning operations. DOD is now looking to industry to apply AI to the hierarchy of data needs. Similarly, IC understands that only a few commercial entities hold LLMs with models in the trillions of parameters, and government access to testing and evaluation is via APIs. This requires new paradigms of government-industry collaboration for large and complex models to ensure ethical and trustworthy AI understanding before adoption into government systems.
Learn about our role as an industry collaborator delivering trusted mission AI solutions
