As we transition from the high-speed content production of 2026 into a more strategic era of talent development, the role of the Learning and Development professional is undergoing a profound transformation. Sofia Khaira, a seasoned expert in diversity, equity, and inclusion as well as workplace technology, has spent years observing how organizations balance technical innovation with the human element of growth. In this discussion, we explore the shift from simply filling training libraries to fostering genuine performance improvements. Sofia provides a deep dive into the evolution of “just-in-time” learning, the necessity of capturing internal expertise, and why the traditional 40-minute module is becoming a relic of the past in favor of real-time, AI-driven support.
Throughout 2026, we have seen AI allow for the rapid creation of massive training libraries in mere minutes, yet many organizations find that employee performance remains stagnant. Why has this explosion of content failed to move the needle on actual skill development?
The reality we faced throughout 2026 is that we treated learning like a volume game, assuming that more content naturally led to more knowledge. While AI cut production times from weeks down to just a few hours, it created what I call a “poisoned chalice” for many HR teams. Employees were suddenly overwhelmed with more modules than ever before, often feeling forced into a culture of “learning for learning’s sake” rather than finding actual solutions to their daily hurdles. When content is nearly free to produce, the quality and relevance often take a backseat, leaving staff frustrated by formulaic courses that don’t address the unique nuances of their specific roles. To see real performance gains, we have to stop measuring success by the size of the library and start looking at whether that information is actually helping a person solve a problem in the heat of the moment.
If the era of the traditional 40-minute training module is fading, how do you see AI effectively integrating learning into the actual flow of daily work?
The shift we are anticipating for 2027 is all about moving away from pulling people out of their work for long, structured sessions and instead giving them answers exactly when they need them. Imagine an employee hitting a roadblock with a complex procedure; instead of waiting weeks for a scheduled workshop or digging through an outdated PDF, they can ask an AI assistant for the specific step-by-step guidance drawn directly from the company’s internal documents. This real-time support allows for a level of agility we simply haven’t seen before, as it provides the right procedure at the exact point of need. Of course, structured courses will still exist for foundational knowledge, but the “everyday” questions will be handled by intelligent systems that understand the specific context of the organization’s unique processes. This removes the friction between “doing” and “learning,” making them two sides of the same coin.
Personalization has been a buzzword for years, but it often just means suggesting different courses. How can AI move beyond basic recommendations to truly tailor the learning experience to an individual’s specific expertise?
The traditional approach to personalization was often superficial, essentially offering different versions of the same generic course, which is both inefficient and demoralizing for experienced staff. In the coming year, AI will allow us to move toward a model where the content itself adapts based on the learner’s existing knowledge and their specific areas of difficulty. For instance, a new starter who is completely unfamiliar with a system will receive detailed, step-by-step guidance, whereas a veteran employee might only receive a quick 30-second update on a specific process change. This prevents “learning fatigue” by ensuring that no one is forced to sit through information they already master, while simultaneously providing a safety net for those who genuinely need the extra support. By mapping content to specific skills, we can finally stop treating our workforce as a monolith and start respecting the diverse experience levels within our teams.
Much of a company’s most valuable knowledge is held by “time-poor” experts who rarely have the bandwidth to write manuals. How can we use technology to capture this “tribal knowledge” without adding to their workload?
This is one of the most exciting shifts we are seeing as we head into 2027, as it addresses the chronic problem of losing vital information when a key person leaves or is simply too busy to help. AI now allows these subject matter experts to simply record a quick voice note or a video of themselves explaining a complex process, which the system then transforms into structured working documents and searchable guides. This democratizes expertise, making the wisdom of a senior manager or a technical lead available to the entire team without requiring them to repeat the same answers dozens of times. However, we must maintain human oversight in this loop; a manager or a trusted peer still needs to verify that the AI-generated documentation is accurate and reflects the most current policies. It’s a partnership where technology handles the heavy lifting of documentation, but humans provide the essential judgment and “sanity check.”
As we move away from tracking simple course completion, what metrics should L&D leaders focus on to prove that their initiatives are actually driving business outcomes?
We are finally moving away from the “check-box” mentality of the past, where a 100% completion rate was the primary goal for an L&D team. In 2027, the focus is shifting toward how quickly a new starter can get up to speed and how effectively they can apply their new skills in a live environment. We need to look at the correlation between learning data and actual business performance—for example, looking at whether a targeted training intervention led to fewer errors in a specific department or faster project delivery times. By connecting these dots, L&D leaders can move from being a “cost center” to a strategic partner that identifies and fills skills gaps before they become operational crises. It’s about impact, not just activity; we want to see that the support provided by AI is actually helping people do their jobs better, rather than just taking up more of their time.
What is your forecast for workplace learning in 2027?
I believe that 2027 will be the year where the “human in the loop” becomes more important than ever, even as our tools become more sophisticated. We will see a move away from massive, generic content dumps toward highly curated, internal knowledge ecosystems where AI acts as a bridge between a question and a reliable, company-specific answer. Success will not be defined by how much AI-generated content a company owns, but by how well they use that technology to foster a culture of curiosity and continuous sharing. Organizations that prioritize the quality and accuracy of their internal data—while ensuring their managers have the “judgment” to know when an AI answer doesn’t quite fit a human situation—will be the ones that truly thrive and see a measurable return on their talent development.
