Ò»±¾µÀÎÞÂë

Lawfully Speaking Returns: Judge Baker Explores AI's Legal Landscape

At a time where technology is impacting all aspects of our lives, from healthcare to warfare, the implications of artificial intelligence (AI) on national security has become a pressing concern. The challenge lies in finding the balance between the limitless potential for AI and the risk of unintended consequences that may impact our privacy, security, and democratic values.

To offer his expertise on this subject, the Honorable Judge James E. Baker joined the Carnegie Mellon Institute for Strategy and Technology (CMIST) for the second installment of Lawfully Speaking: A Forum on Law and Ethics. His lecture, entitled Promise and Peril: The Next Ten Years of National Security AI and its Regulation, addressed the future of artificial intelligence (AI) regulation and national security. A tenured professor at Syracuse University College of Law with a courtesy appointment in the Maxwell School of Citizenship and Public Affairs, and Director of the Syracuse University Institute for Security Policy and Law, Judge Baker currently teaches courses on national security law, emerging technologies, ethics, and related subjects. He also served as a Judge and Chief Judge on the U.S. Court of Appeals for the Armed Forces and as a Marine Corps infantry officer. 

In her introduction, CMIST Director, Audrey Kurth Cronin, emphasized Judge Baker’s commitment to transparency as well as how his career demonstrates the importance of public service both in the military and in the broader context of law.

Building upon Cronin's introduction, Judge Baker began his remarks by weaving together seemingly distinct aspects of his career, such as how his experience serving in the Marines, including the need to make and be accountable for many decisions, helped prepare him for practicing law. He stressed the importance of technology policy and law, especially with AI, and described the present moment as “the most challenging time in the field [of national security] in the span of [his forty year] career” in large part due to the world of emerging technologies. 

Much of Judge Baker’s role includes translating the general strengths and weaknesses of AI into plain language in order to meet the needs of policymakers and thus enable them to take advantage of all available tools. He stressed the importance of ensuring all parties are discussing the same technology or issue when engaging in conversations about AI, highlighting the need for standard definitions. Practicing his own advice, he provided the audience with the , “the ability of a computer system to solve problems and to perform tasks that have traditionally required human intelligence to solve.” Given the national security imperative posed by AI, Judge Baker emphasized that applying law and ethics leads to better AI outcomes, limits bias, and achieves more accurate technology. He stated, “AI is and will be used in all aspects of national security,” and acknowledged that this reality offers many helpful tools but also carries many risks, such as potentially increasing, instead of decreasing, conflict.

Judge Baker raised three important considerations regarding the future of AI. The first, the Centaur’s Dilemma, revolves around the balance between AI capacity and human control. The more instantaneously AI is able to act, the less human control there is. However, as human control increases, AI capacity becomes delayed and thus its benefits may decrease. He stressed the need to locate the human control aspect, which is always present, in each AI use case. Second, Judge Baker emphasized the need to understand all of the types of bias that can get embedded in an AI system and then address them. Third, Judge Baker underscored the critical role of data quality in AI technology, noting that the data used for training AI models must be carefully vetted to ensure accurate and objective outcomes.

Looking towards the next ten years of national security as it relates to AI, Judge Baker walked the audience through a variety of national security use cases, such as turning to AI in situations characterized by “fear, fatigue, and repetition.” Judge Baker addressed concerns about the use of AI in these scenarios, such as the implications of AI autonomy, access to data, and the dangers of low bars to entry. He also presented a toolkit for approaching AI governance, which included arms control tests, regulatory thresholds, and wargaming.

While AI, due to its innovative nature, sometimes poses new questions for U.S. law, there are resources within our legal tradition to help address these challenges. For instance, Judge Baker explained that the process of applying existing law, by utilizing the concepts of law by analogy or metaphor, can help bridge the gap until new laws more specifically applicable to AI are created. 

Judge Baker challenged audience members to actively engage in shaping AI governance by making purposeful decisions in how they use the technology. He stressed the need to clearly define AI models and their intended purposes, to consider the balance between human and AI understanding, and to hold the government accountable for its use and governance of AI. He urged audience members not to let lack of mastery stand in their way, but rather to "Spot the issue, don't be the expert on all things. Then go find the expert." 

Furthermore, Judge Baker emphasized that in U.S. policy and legal debates, individuals should take a proactive role in determining the circumstances and methods of AI deployment, rather than passively claiming that they cannot employ it. With implications that reach far beyond this talk, Judge Baker called to those in attendance to continue shaping the future of AI and national security, “We can sit back and watch all of this and hope for the best, or we can shape the outcomes.”

(Image 1: CMIST Director Audrey Kurth Cronin offers introductory remarks; Image 2 left to right: Judge James E. Baker and Professor Audrey Kurth Cronin; Image 3: Judge James E. Baker; Image 4: Attendees listen as Judge Baker addresses the future of AI regulation and national security; Image 5: Judge Baker's lecture was titled "Promise and Peril: The Next Ten Years of National Security AI and its Regulation"; Image 6: Judge Baker engages the audience)