By Mickey Clark
Every conversation about artificial intelligence (AI) in learning and development (L&D) carries both excitement and apprehension. But one fact remains clear: without strong learning foundations, AI will struggle to deliver on its promise.
Before learning can become adaptive, it needs something solid to adapt from. This means clear structure, consistent frameworks, and integrated systems that support data-driven insight. Adaptive learning cannot function in isolation; it depends on the reliability of what comes before it.
“Adaptive learning is the roof of a well-built house. Without strong foundations, even the smartest systems will wobble.” ~ Mickey Clark
Laying the Groundwork
Three fundamentals help determine whether adaptive learning will thrive in the workplace.
1. High-quality, structured content
Adaptive learning systems rely on clean, modular content to determine what to serve and when. If the source material is inaccurate, inconsistent, or poorly organised, AI will only amplify those flaws. The old saying “garbage in, garbage out” remains true, only faster and at scale.
As content becomes more dynamic, a well-defined learning taxonomy is essential for both humans and machines to navigate effectively. Recent findings from the World Economic Forum’s Future of Jobs Report (2024) reinforce this need, noting that organisations that invest in structured, high-quality learning data are better positioned to reskill employees rapidly and keep pace with AI-driven change.
2. Clear competency and performance frameworks
AI cannot personalise learning if it does not understand what success looks like. Competency maps outlining the skills, behaviours, and performance levels that matter most form the backbone of adaptive systems. When these frameworks are incomplete or outdated, even the most advanced algorithms can misinterpret what a learner needs next.
3. Systems that speak the same language
Learning management systems (LMS), learning experience platforms (LXP), and authoring tools all depend on compatible data standards such as xAPI and cmi5. According to the Advanced Distributed Learning (ADL) Initiative, these standards are designed to ensure interoperability across learning environments, allowing data to move freely between systems and enabling richer insights into learning effectiveness.
While Australia does not yet have a national framework equivalent to the US ADL’s “Total Learning Architecture,” the ADL standard demonstrates the level of precision required for true interoperability. It defines how learning systems, content repositories, and analytics services should interact; an important benchmark for Australian organisations seeking to future-proof their learning ecosystems.
When data flows cleanly, L&D teams can see how learning impacts performance, creating a vital foundation for adaptive intelligence.
When Foundations are Firm
Once structure, clarity, and data interoperability are in place, adaptive learning has a better chance to shine. Supportive AI tech is able to adjust the depth, pacing, and difficulty of content to match each learner’s role and readiness. These foundations finally enable one of L&D’s most sought-after goals: personalisation at scale.
A Practical Example: Adaptive Onboarding
Traditional onboarding often follows a fixed schedule of information and assessments, regardless of prior knowledge. An adaptive system, by contrast, begins with a quick diagnostic and tailors the sequence of learning. It may skip what a new employee already understands, dive deeper where gaps appear, and even vary content format to sustain engagement. The result is faster confidence, earlier contribution, and greater job satisfaction.
Learning from the Past
If all the thoughts about interoperability and getting the foundations right might sound a little abstract, an example from the Industrial Revolution may offer a fitting analogy.
When textile mills first adopted mechanised looms in the 18th century, many struggled, not because the machines were flawed, but because their infrastructure and workflows were unprepared for mechanisation. Historical accounts, such as those in The Age of Manufactures (Berg, 1994), note that early power looms required stable foundations, precise alignment, and skilled operators before they could run efficiently. The same holds true for AI-driven learning: without robust content, frameworks, and systems, the technology’s potential remains underused.
Preparing for What’s Next
As AI continues to reshape learning, L&D professionals are uniquely positioned to ensure that human understanding drives technological adoption, not the other way around. By strengthening learning foundations now, organisations can create adaptive ecosystems that are intelligent, resilient, and ready for whatever comes next.
Further Reading and Resources
- Berg, M. (1994). The Age of Manufactures, 1700–1820: Industry, Innovation and Work in Britain. Routledge.
- World Economic Forum. (2024). Future of Jobs Report.
- Advanced Distributed Learning Initiative. (n.d.). xAPI and cmi5 Standards Overview. Retrieved from https://www.adlnet.gov/guides/tla/service-definitions/
About the Author: Mickey Clark
Mickey Clark is Co-Founder and Chief Operating Officer at RockMouse, a digital learning innovation consultancy helping organisations design adaptive, human-centred learning ecosystems that deliver measurable performance impact. Mickey has a long career in business and education, specifically in technology-related business solutions, and has provided learning solutions to most of Australia’s blue-chip corporates and government agencies, federal and state, through his company, The Learning Group. Prior to that, Mickey served as an International Director at Dun & Bradstreet Software, serving in a variety of R&D, marketing, and product management roles, before being chosen to head up the company’s Product Development and Support Division in Japan and Asia Pacific.
