AI has become one of the government’s most powerful productivity weapons and has untapped potential as a training tool. The United States military is using AI to deliver personalized scenario-based training programs that adapt to each user’s learning progression and experience levels. While some agencies are using AI to identify skills gaps, there is a lot more that they could be doing when it comes to using AI to drive learning.
By analyzing individual skill levels and learning behaviors, AI can dynamically tailor educational content, pacing, and support in real time. The result is a deeply personalized learning journey that goes beyond what traditional classrooms or even one-on-one instruction can offer.
Here are four ways agencies can use AI to create customized learning experiences for all employees.
AI Adaptive Training
Adaptive training programs adjust course material and instruction based on multiple factors such as an individual’s performance, behavior, skill level, and learning preferences. For instance, an AI-based adaptive training program might assess a person’s capabilities based on the accuracy of their answers to a quiz and adjust questions or training accordingly. A warfighter participating in a simulated tactical scenario that changes based on their decisions is an example of adaptive training.
Adaptive training involves a combination of machine learning (ML) and data analysis. These processes model the behavior patterns of users and feed that information back into the AI to personalize their learning paths.
With adaptive training, agencies can create more dynamic and effective programs than traditional “one-size-fits-all” approaches.
Multimodal AI Training
Some people learn better through visuals or sounds; others through words; still others through a combination of inputs. Multimodal training uses multiple types of input methods, including text, imagery, video, and sound, to train users. For example, a person might learn via a combination of text-based quizzes, instructional videos, or voice instructions. Multimodal training can be adaptive, with AI customizing delivery methods depending on the situation and the user’s preferences.
Augmented virtual platforms are especially effective for simulation-based training. These AI-powered environments support situational learning by allowing users to see, hear, and even interact with potential outcomes based on their choices, creating a more immersive and realistic training experience.
Multimodal training significantly expands educational access for individuals who face challenges with traditional formats. For example, natural language processing (NLP) tools such as speech-to-text and voice-based interfaces can be invaluable for visually impaired users or those with limited motor skills.
Personalized Training at the Edge
Much of the public sector operates in highly distributed settings, from warfighters in harsh, low-connectivity environments to government employees in remote offices. It can be challenging for agencies to implement in-person training across these various locations.
Advances in compression technology have enabled the downsizing of large language models into more efficient small language models that can run locally on individual devices. This shift allows AI to function effectively in air-gapped or low-connectivity environments without depending on cloud infrastructure or centralized inference engines, resulting in security and cost advantages.
Local AI models can deliver personalized learning experiences directly at the edge by analyzing real-time human performance data and dynamically adjusting instructional content. This enables tailored skill development, even in remote or secure environments.
By using smaller, task-specific models trained on curated, high-quality data rather than massive, generic datasets, agencies can further adapt learning to the specific needs of roles or missions. This approach makes training both more relevant to individuals and more scalable across a distributed workforce.
Training Opportunities Through Prompt Engineering
Agencies can engineer AI prompts to test for organizational skills gaps. To create effective prompts, an agency might start by defining key mission areas or emerging technology domains, then ask the AI to compare those priorities against available training records, certification levels, or performance reviews. By tailoring prompts to specific datasets and objectives, agencies can guide AI to reveal patterns, flag critical gaps, and recommend areas for targeted upskilling.
For example, the DoD might prompt AI to analyze soldiers’ proficiency in drone navigation and identify areas for training and improvement. Based on this data, the organization may create a customized drone training program. Training can then be personalized and adapted to the skill levels and learning preferences of individual soldiers.
As agencies build out their AI roadmaps over the next few years, they should consider how to integrate the technology into their training programs. The four use cases outlined here are a good starting point and exemplify the technology’s potential as the most adaptable training tool ever available to the public sector.
Burnie Legette is a Solution Architect and Specialist for AI and Data Operations at Intel
Leave a Reply
You must be logged in to post a comment.