AI and the Bucking Bronco
Many organizations are still trying to stay in the saddle. The more interesting opportunities may emerge after AI becomes an ordinary part of work.
A few years ago, AI occupied a strange place in organizational conversations. People could see something powerful taking shape, but few seemed confident about what it would become. Demonstrations alternated between astonishing and disappointing. Predictions swung wildly between transformation and irrelevance. Most discussions carried an underlying assumption that something important was happening, even if nobody could quite agree on what that meant.
People who work with me know I tend to think in analogies. Lately, whenever AI comes up, I find myself thinking about a wild horse on the horizon. At first, most organizations weren’t entirely sure what to make of it. They could see there was something powerful there, but nobody knew exactly how useful it would be or what it would eventually become. Over time, the conversation shifted from whether the horse mattered to whether it could be ridden. Some organizations are still deciding whether it’s worth approaching. Others have already climbed into the saddle and discovered that staying on a bucking bronco is considerably harder than it looks. A smaller group has begun figuring out how to direct the horse rather than simply hang on.
That progression mirrors much of what has happened with AI. Organizations that initially viewed the technology as an interesting curiosity now treat it as a strategic priority. Some have developed repeatable workflows and meaningful use cases. Others continue to struggle with inconsistent results, questionable outputs, and the gap between what AI demonstrations promise and what day-to-day work actually requires. The conversation has nevertheless shifted. In many organizations, the debate is no longer whether AI matters but how it should be used, where it creates value, and what role it will eventually play in the work itself.
For many organizations, that remains the primary challenge. Learning how to work with AI is still difficult, results remain inconsistent, and governance questions remain unresolved. The distance between a demonstration and a scalable business process is often larger than it first appears.
Most technologies eventually become ordinary. Search engines, smartphones, cloud computing, and video conferencing all passed through periods when they were discussed constantly. Over time, attention shifted away from the technology itself and toward the ways it changed behavior. Nobody builds a strategy around using email. Organizations build strategies around communication. The technology becomes infrastructure.
At some point, the ability to use AI effectively may no longer distinguish an organization from its competitors. Most organizations will get there eventually. Some will arrive sooner than others, but proficiency has a habit of spreading. The techniques that seem innovative today will become commonplace, and many of the advantages organizations currently celebrate will eventually be viewed as standard operating practice. The horse will still be useful, but it will no longer be the most interesting thing in the field.
For decades, many learning teams have operated under a fairly straightforward assumption: expertise is scarce, and part of the organization’s job is distributing it. Courses, workshops, documentation, and knowledge bases all exist because information needs to move from people who know things to people who need to know them.
An employee struggling with a software process no longer needs to wait for the next training session. A manager preparing for a difficult conversation can generate examples, talking points, and coaching suggestions on demand. A new hire can ask follow-up questions without worrying about exhausting the patience of a colleague. None of these interactions replace experienced professionals, but they do begin to alter the relationship between people and information.
Much of the current AI conversation in learning and development focuses on efficiency. Tasks that once required hours of effort can now be completed in minutes, allowing learning teams to move more quickly and respond to requests that might previously have exceeded their capacity. Those improvements matter, particularly for departments that have spent years balancing growing expectations against limited resources.
Yet speed has a way of dominating the conversation. A horse can get someone to a destination more quickly than walking, but that does not automatically make walking obsolete. People continue to walk because it serves purposes that have little to do with transportation. The same may be true of some forms of learning and work, where efficiency matters but does not necessarily capture everything that makes an activity worthwhile.
If employees have easier access to information, perhaps the role of learning shifts toward application rather than acquisition. If explanations become abundant, perhaps practice becomes more valuable. If knowledge can be retrieved instantly, perhaps judgment becomes a more important differentiator than recall.
Instructional designers may spend less time creating content and more time designing experiences. Learning teams may devote more attention to performance support and less to information delivery. Managers may become more central to development efforts as coaching, feedback, and workplace application grow in importance. The questions themselves may change from “How do we teach this?” to “How do we help people use this effectively?”
The strongest AI users I know spend surprisingly little time talking about AI itself. They have already accepted that learning to work with the technology is part of the job. Their attention has shifted toward workflow design, organizational capability, decision-making, performance support, and the changing nature of expertise itself. They also seem less interested in claiming credit for every idea that passes through their hands. As AI becomes more capable, professional credibility may depend less on presenting ourselves as uniquely brilliant and more on demonstrating sound judgment. Knowing how to use the tool is becoming increasingly important. Understanding what the tool contributed may be just as important.
Right now, most organizations are still occupied with the bronco. Some are deciding whether it’s worth approaching. Others are learning how to stay in the saddle. A smaller group has become remarkably effective at directing the horse toward productive work. All of that effort matters, but what remains uncertain is what learning, work, and expertise will look like once riding becomes ordinary.
Much of today’s AI conversation sounds like a society trying to create better riders. That work is necessary. The horse still needs to be tamed. At some point, however, transportation stops being a story about horses. For learning and development, the more interesting challenge may be recognizing when that moment arrives and what comes next.
