The Centaur’s Dilemma:
National Security Law for the Coming AI Revolution
The eye-opening moment of The Centaur’s Dilemma comes when you realize you are not only gaining a profound perspective into National Security Law but you also learn how AI implications are in almost every legal practice area.
It's hard to have a complete understanding of … what decisions are actually being made algorithmically and which are being made by people …. When you don't have that, I would argue you have the risk of no longer having real understanding or control of your organizations.
~ General (Ret) Stanley McChrystal
Asking a lot more questions than it answers, The Centaur’s Dilemma is a must-read for any lawyer interested in gaining a better understanding of where artificial intelligence (AI) could impact their practice of law. Mr. Baker provides volumes to think about, questions to begin asking, and takeaway points at the end of each chapter to improve retention. The eye-opening moment of The Centaur’s Dilemma comes when you realize you are not only gaining a profound perspective into National Security Law but you also learn how AI implications are in almost every legal practice area.
In what is described as a “defining technology of the twenty-first century … China’s state council announced a $150 billion centralized program to develop AI and become the world’s leader in AI by 2030 … [and] Vladimir Putin declared ‘whoever controls [AI] will be the ruler of the world.’” Mr. Baker does not provide a similar bold statement from our policymakers. Of potential concern, he lets us know that our policymakers “understand the importance of operational timelines.” However, “… they are less conscious of the risks of acting at machine speed ….” Therefore, he lays out that currently government, private corporations, and academia are making their own policy rather than working toward a unified policy for national security. “If the government does not act, private actors and litigation will play a disproportionate role in defining national policy.
China’s state council announced a $150 billion centralized program to develop AI and become the world’s leader in AI by 2030.
Perhaps drawing on his time as a judge on the United States Court of Appeals for the Armed Forces, Mr. Baker categorically sets out a legal framework moving you through Constitutional issues. The Fourth and Fifth Amendment concerns for attorneys include understanding potential data bias and that “AI-generated results will serve as a predicate for probable cause ” as well as Sixth Amendment concerns on whether the defendant will be able to “question the author of the algorithm.
That said he also touches on contract/fiscal law, weapons reviews, labor law, and ethics—asking questions such as: Does the Federal Government have sufficient numbers of qualified attorneys to be able to advise decision-makers during the potential collection, storage, use, and buying of data? Moreover, if a new weapon or weapons system incorporates AI, DoD Directives state that attorneys will need to do a legal review on acquisition and procurement as well as compliance with domestic law, treaties and international agreements. Do we have personnel capable of doing these reviews and doing so at the R&D stages?
In Chapter 10, Mr. Baker turns to ethics. The Centaur’s Dilemma makes it clear “lawyers have a professional responsibility to understand and identify the ethical questions and dilemmas associated with AI-enabled systems and machines .… The lawyer who is not conversant with AI will not be invited into the decision making room and will not hold their place in that room.” They also will not be “consulted at the research and development stage of AI system creation.” If brought in at all, lawyers will be at the use stage with “fewer opportunities to ask the right questions and even fewer opportunities to guide AI applications to preferred outcomes ….”
I highly recommend The Centaur’s Dilemma to all practicing attorneys but especially government attorneys. As mentioned above, AI affects contract/fiscal law reviews. Administrative law attorneys will need to be conversant in AI when they review investigations and work with digital forensic experts to analyze audio and video evidence. Labor counsel need to understand AI when it is part of the hiring process. National Security Law advice will cover questions on sending out “the perfect wingman” or relying on AI outright to complete a mission. Finally, litigators will need to know who was involved at each stage of AI development, how to understand the data sets, and was the end-use reliable. Once we can answer these questions, we will be able to determine whom, if anybody, should be held accountable for what happened. The bottom line is that Mr. Baker is correct: AI will be the “defining technology of the twenty-first century.” It will affect every aspect of our legal practice and as The Centaur’s Dilemma makes abundantly clear, we need to be ready.
EXPAND YOUR KNOWLEDGE
External Links to Additional Resources
About the Author
Lieutenant Colonel Timothy D. Litka, U.S. Army
(B.A., University of Akron; J.D. University of Toledo College of Law; LL.M., The Judge Advocate General’s Legal Center and School) is currently assigned as the Senior Legal Advisor and Instructor at the United States Army’s Command and General Staff School, Fort Leavenworth, Kansas. He is licensed in the District of Columbia.
James E. Baker, The Centaur’s Dilemma, National Security Law for the Coming AI Revolution,
(The Brookings Institution 2021).