The U.S. Navy has moved ahead in developing and fielding unmanned systems on, above, and under the ocean. Navy leaders envision a future in which manned and unmanned ships will sail side by side, and unmanned systems will operate over the horizon supported by shore-control facilities. The Navy is funding research and development for 215 projects related to artificial intelligence (AI), including greater autonomy for unmanned systems and aiding warfighter decision-making. Despite fielding physical systems, the Navy has not defined the training and education for operators and maintainers of AI systems. This despite a 2020 plan to educate and train Department of Defense (DoD) employees. The Department of the Navy (DoN) has released several strategy documents about AI and autonomous systems including the Strategy for Intelligent Autonomous Systems and the Unmanned Campaign Framework. Unfortunately, these documents both avoid discussion of how the Navy will train sailors or operate future unmanned systems.
The Navy has been slow to establish the necessary career frameworks for sailors that operate unmanned systems, while losing talented sailors to the private sector where their skills are in demand. As AI-enabled naval systems move from the lab to the battlefield, an educated and trained workforce must be ready to meet them.
Why the Navy Needs AI Talent in Uniform
For several years, across multiple defense strategy documents, the DoD has acknowledged the role that AI will play in the near and distant defense systems. The 2018 National Defense Strategy identifies advanced autonomous systems as a key modernizing capability and states that the DoD must broadly invest in these technologies. The 2020 DoD Education AI Strategy stated that the department must develop "world class" AI practitioners to make AI a reality within DoD. The National Security Commission on AI's final report recognizes the importance of talent, stating "The AI competition will not be won by the side with the best technology. It will be won by the side with the best, most diverse, and tech-savvy talent. The DoD . . . face[s] an alarming talent deficit." The report recommends, among other actions, that digital talent must be organized into individual corps and that those within DoD with requisite digital talents should be able to spend a career within this specific career field.
The personnel with the skills to build and manage AI systems must have defined and approved career tracks that keep them in relevant jobs and provided them with the tools and continuing education to succeed for the duration. Those with the skills and motivation to work on AI projects will not want to spend several years on sea duty, away from the cutting edge of AI, to meet an undesired career goal
Just as the Navy currently requires basic training and literacy in cyber security, so too it must require basic training in AI. The Joint Artificial Intelligence Center (JAIC) has already begun the work to develop and pilot AI education courses for a general audience. Since AI education will happen across DoD, the Navy will not have to develop course content on its own. Instead, the Navy can identify where in a sailor's initial and continuing education AI education courses fit and ensure that training is delivered.
Increased adoption of AI will automate many tasks that humans perform today, in both offices and on the battlefield. The changes from adoption of AI systems will fall disproportionately on career fields that predominantly perform administrative and logistical tasks. Intelligence specialists, who pour over images and documents looking for ships or aircraft, will soon be assisted by AI systems that can perform the same tasks faster and more accurately.
The Navy of the near future may have fewer jobs in some career fields and more in others, and automation can spur demand for work when automation makes that work cheaper and faster. The daily tasks that sailors perform may involve more time supervising AI systems than the manual tasks they currently perform. It is up to Navy Personnel Command to match current career fields to these archetypes and identify how they will change.
DoD recognizes that it cannot compete with the private sector in hiring technical talent for AI projects. However, the Navy will need sailors highly educated in AI to manage the acquisition of AI systems, develop AI system requirements, and work with Navy civilian employees to identify new AI research areas to benefit the fleet.
Fortunately, DoN does not have to exclusively hire this talent from outside sources. There are many serving in uniform today with the skills, often acquired at DoN expense, to work on AI projects within the Navy. These sailors comprise a "hidden AI workforce." There are many current career fields, both military and civilian, that are either directly related to AI or AI-adjacent. Service members with a passion for this type of work and with a desire to see reform within DoN end up working on these passion projects on their own time when they could be adding significant value to DoN by working on these projects full time. These sailors have the skills, the Navy must provide them with the opportunity to use these skills to create AI systems that can operate on the battlefield.
Many of the sailors within this "hidden AI workforce" received their skills at Navy expense. Naval Reserve Officer Training Commands and the U.S. Naval Academy commission new ensigns every year with STEM degrees in computer science, math, operations research, and other disciplines relevant for AI work. The Navy pays for master's degrees in similar fields at the Naval Postgraduate School and civilian universities. Officers who earn these degrees often return to fleet units to with little application of their advanced education. These sailors represent a cadre of talent that the Navy can employ in AI projects without expending the time and money to recruit outside talent.
Both the Army and Air Force have partnered with the Massachusetts Institute of Technology and Carnegie Mellon University to accelerate AI innovation and bring defense problems to civilian researchers. These arrangements provide active-duty officers and civilian employees with advanced education, allowing them to return to relevant jobs within their departments to lead AI projects. These units provide a means for the services to grow their pool of digital and AI talent from within. The Navy should seek to build these partnerships with academia to leverage both research and education resources.
Without a doubt, the goals defined here will take many years of effort across the Navy to implement. However, there are some steps that can be taken with relatively little cost and reorganization. First, the Navy should develop an AI education implementation plan based on the 2020 DoD AI Education Strategy. This plan would ensure that the Navy is meeting the objectives of the DoD strategy without duplicating efforts currently underway in other offices like the JAIC.
Second, the Navy can work to understand the current capabilities of its workforce. A review of personnel records for relevant educational backgrounds and survey of related subspeciality and career codes provide a starting point. Surveys of sailors in technical career fields can also identify those with the talent and desire to work on AI systems. This is the first step to identifying sailors with the skill set needed for AI-related jobs.
Longer time frame tasks, such as the creation of AI training programs, must begin as pilot projects before scaling up. An academic partner institution can establish a Navy AI Center of Excellence like with the Army and the Air Force. This partnership would bring relevant Navy datasets and problems to academic AI researchers and provide mid-career officers and senior enlisted with advanced education in the field of AI.
If the Navy fails to adapt its talent management systems and careers to current technology, sailors with the knowledge and passion for unmanned systems will leave the Navy. Watchstanders ashore and at sea will not fully grasp how to interact with and operate AI systems. Most of all, the Navy may find itself divided into uniformed operators stuck in their largely traditional role, and civilian contractors who maintain and build unmanned systems without informed input from uniformed sailors. Failure to adapt its workforce and processes for AI adaptation will leave the Navy unprepared for great power conflict this century.