“We can create the course in 6 minutes!” My friend explained this with mixed pride and uncertainty, unsure of if this was good or bad. Her vendor demo, like many I’ve seen, showed the speed that AI tools can create training. Many hail generative AI tools, which promise to revolutionize training and education, as the next big thing. However, beneath the hype lies a concerning reality: these tools often claim to be quick, yet they fall short of creating learning experiences that are actually effective.f
Misunderstanding Effective Training
The driving force behind the shortcomings of generative AI tools is a combination of technology companies and the business world, neither of which truly understands what makes for effective training, despite thinking otherwise. While technology companies excel at software development, they lack a comprehension of modern learning principles. As a result, the solutions they offer often perpetuate outdated and ineffective training models.
Likewise, business leaders and decision-makers often believe they “understand training”, leading them to seek out or be captivated by flashy features in courses, all while failing to recognize the elements that truly drive behavior change in effective learning solutions. I’ve witnessed firsthand how leaders often instruct their L&D teams to “save money by using AI tools to create training,” without considering that these tools might not actually save money, especially if the AI-generated training turns out to be ineffective.
The Flawed Perception of Effective Training
Technology companies venturing into the L&D space tend to view training as a straightforward process. They believe that if they can automate the creation of “level 1 AI” like quizzes, slides, video creation, image generation, and other learning assets, they’ve solved the problem. However, effective L&D requires much more than this. It involves understanding the learner persona, how adults learn, and how to design solutions that promote long-term behavior change.
A business leader not in L&D recently told me how excited he was that his team could now use AI to create training slides with a prompt, complete with speaker notes to be read. What he doesn’t understand is that the AI-generated text-heavy slides, while well-phrased and visually appealing, lack cohesion. And even if they were coherent, lengthy lecture-based training is not effective for learning transfer.
Often, AI solutions automate draconian teaching methods from the past, like quizzes for remediation and lecture-style videos. This approach does little to promote genuine learning or skill development.
The Generic Trap of AI-Generated Training
Another fundamental issue with common generative AI L&D use is their reliance on generic data sets. These tools pull information from the internet, a vast repository of content. While this might seem like a strength, it often results in a sea of generic training materials. Whether it’s a course on communication skills or an onboarding module, the AI-generated content tends to be broad and lacks specificity.
Normally L&D conducts a needs analysis process and identifies a unique workplace issue that the L&D team can then examine and derive the best solution for. If performance support or training addresses the issue, effective learning solutions are then tailored to the unique needs of the learners and their work environments.
When content is derived from a single, homogeneous source without a proper needs assessment, it’s unlikely to align with the unique challenges or learning goals identified during the needs assessment. Take my recent experience with an AI-powered simulation, for example. Although the tool itself was impressive, the scenario it generated for “communication skills” was generic, much like countless other examples on the internet. It literally felt like a fictional scenario I’ve read many times on countless business blogs. It failed to target the very specific communication-related performance issue I’d identified as critical to address. Without this focus, the training lost its relevance and didn’t meet the performance objective.
Bypassing L&D is a Risky Shortcut
An increasing concern is that the rise of AI tools has also made it easier for business teams to bypass the L&D department and create their own training. At first glance, this might seem like a cost-effective solution, especially when budget and resources are tight. However, the consequences can be detrimental to the quality of training.
Without the guidance of L&D professionals, who understand instructional design and modern adult learning theory, these DIY training solutions often miss the mark. They may deliver information, but they don’t lead to meaningful learning or behavior change. When business teams bypass the L&D department, it not only undermines the role of L&D professionals but also jeopardizes the effectiveness of training programs across the organization
Speed Over Substance
Finally, the allure of generative AI tools often centers around speed. Vendors promote the idea that training can be created faster than ever before, with some boasting about the ability to generate entire courses in a matter of minutes. While speed can be valuable in some contexts, it’s not the ultimate measure of success in L&D.
Effective training takes time to design, develop, and test. It involves ensuring that the content isn’t just “engaging and interactive” but that it’s actually achieving the performance objective. Skilled instructional designers know how to ignite behaviour change and learning retention, e.g. through intentional video design rather than a talking head video. If the focus is solely on speed, the resulting training may be shallow and ineffective, doing more harm than good in the long run.
A Shift Towards Effective Training with AI
The rise of generative AI in L&D carries a significant risk of creating ineffective and poor learning solutions. The technology companies driving this trend often lack a deep understanding of L&D principles, leading to solutions that are generic and superficial. Moreover, the ability for business teams to create their own training without the guidance of L&D professionals poses a threat to the quality and effectiveness of learning programs.
While speed is often touted as a selling point, it’s not the ultimate goal of training. To ensure that generative AI tools contribute positively to L&D, the focus must shift from speed to effectiveness, emphasizing meaningful learning experiences that foster behavior change and skill development. By doing so, organizations can harness the power of AI without compromising the quality of their training programs.
Recent Comments