Rethinking workforce development: A new approach enabled by AI
Learning and development leaders face increasing pressure to upskill and reskill the workforce more quickly, cost-effectively, and at scale—while keeping pace with evolving skill demands and shifting organizational priorities. The central challenge is not a lack of learning options, but determining when generic content is sufficient and when tailored, organization-specific learning is essential to performance.
In this context, advances in AI are changing the calculus. By reducing the time and cost required to design high-quality, customized learning, they enable L&D teams to rethink long-standing build, buy, and borrow tradeoffs—and adopt approaches that more directly translate learning investments into business and mission outcomes.
Today’s reskilling strategies
Organizations have made real progress on upskilling and reskilling, but each dominant approach carries structural tradeoffs:
Buy (Curated Learning)
Curated courses from third-party learning platforms deliver speed and scale but sacrifice the organizational context that enables optimal on-the-job application.
Borrow (User-Generated Content)
Content created by high performers provides job and organization-specific context but may lack the instructional design rigor that leads to the best learner outcomes.
Build (Original Development)
Custom training delivers instructionally-sound, organization-specific learning but can be time- and resource-intensive to develop.
Thus, the ambition of offering current, organizationally-relevant learning that maximizes learner outcomes in support of mission objectives has required a challenging balancing act of build, buy, and borrow decisions for L&D teams.
Rethinking when to build
Most organizations recognize the value of customized, mission-specific learning, but historically reserve build decisions for only the highest-priority skill gaps.
L&D leaders address the remaining needs through other methods, accepting the tradeoff of more generic learning content in the interest of scalability at lower cost.
L&D teams that embrace AI capabilities and adopt a “build-first” mindset are more likely to meet today’s upskilling and reskilling imperative in a way that translates to business and mission outcomes.
Generative AI requires that L&D leaders revisit this calculus.
Organizations can safely leverage this technology to generate multi-media content—text, image, and even video—in a matter of seconds, and at relatively low cost. Of course, human reviews are still essential to ensure accuracy and instructional quality, and to mitigate risks. So, building new, production-ready courseware will not and should not happen instantly. But it can happen faster than ever before.
When customized, instructionally-sound learning becomes the default instead of the exception, learners will feel even more confident applying their newly acquired skills. They will know what they have learned is relevant to their job and fits within their organization’s policies, processes, and norms. This shift makes it feasible for L&D teams to adopt a build-first mindset that connects learning investments more directly to business and mission outcomes.
How generative AI reshapes instructional design
Traditionally, instructional design teams have followed waterfall approaches to training development—moving from outline to storyboard to script in a linear sequence. Because the cost of rework increases as development progresses, these steps are often tightly structured and iterative, with multiple rounds of review and revision to mitigate risk and ensure quality. While effective, this approach can be time-consuming.
To improve efficiency, many teams have shifted toward more agile ways of working, borrowing from practices in software development. ICF, for example, has used an Agile Instructional Design (AID) model for years. By breaking work into smaller, iterative sprints, teams have achieved meaningful gains in speed and flexibility.
Recent advances in AI enable a fundamentally different approach—further accelerating time to delivery.
Consider two scenarios:
- In the first scenario, you’re asked to review a training outline to see if it will meet the desired learning objectives. The outline provides the skeleton of the lesson with several descriptive bullet points to help you envision how the course will flow.
- In the second scenario, your charge is the same—to see if the learning solution will meet the desired learning objectives. But this time, you are given an 80% learning solution, complete with a fully drafted narrative script, visual imagery, or even video.
Scenario two is, of course, preferred. You get a better sense of the content, look, feel, and tone of the learning session right from the outset. This more complete picture allows you to provide more constructive feedback and to reach a shared understanding of what good looks like much more quickly.
Scenario two was previously impractical. Producing even a partial solution required weeks of effort, with additional time to incorporate feedback. Today, AI-enabled content generation and iteration allow instructional designers to present more complete drafts and move from concept to final product in a fraction of the time.
For L&D leaders, the implication is not simply faster production, but earlier clarity on whether a learning solution will deliver the intended outcomes.
Applying a build-first approach
Scenario two is no longer theoretical. Generative AI is already being used to support learning design and development at scale. ICF has applied text, image, video, and voice-generation technologies within secure environments to accelerate e-learning production for public and private sector clients, while maintaining appropriate human review and quality controls.
Generative AI does not simply accelerate learning development—it reshapes how organizations think about workforce capability. By making high-quality, customized learning faster and more cost-effective to produce, it allows L&D leaders to shift from rationing build decisions to using them strategically where context and performance matter most. Organizations that adopt this build-first posture, supported by responsible AI and human oversight, are better positioned to scale critical skills, reinforce organizational ways of working, and translate learning investments into sustained business and mission outcomes.