4 years back, Dan, an Instructional Designer, spent 6 weeks designing a compliance training module. Late nights. Endless review cycles. Stakeholder feedback that rewrote everything in week 4. Fast forward to today… AI can build that same module in an afternoon. GenAI has cut eLearning development time by up to 70%. The content is cleaner, faster, voiced, structured. And L&D teams everywhere are celebrating the efficiency. But here's what the data is quietly revealing: learning outcomes haven't kept pace. We have more content than ever. But we have the same retention problems, or even worse. AI accelerated the part of instructional design that was never actually the problem. The broken part was understanding why people don't apply what they learn, designing for how the brain actually retains information, connecting training to a real behavior change that still needs human thinking. Deep, slow, deliberate thinking. The teams getting this right are asking different questions before a single module is built: not "what content do we need?" but "what behaviour needs to change, and what's stopping it?" That second question is what separates good learning from forgettable training. Great instructional design doesn't just deliver information; it anticipates the real world the learner is walking back into. The habits they'll revert to. The moments where the new skill will feel uncomfortable. And it designs for those moments, not just the classroom. That shift, from content design to performance design, is where the real work lives. And it's work AI can support, but not replace. Working with organizations across industries, we've seen the teams getting real results aren't the ones moving fastest. They're the ones using AI's gift of time to go deeper into the design, not wider into the content library. The bottleneck was never production. It was always the design. Your team just got 70% of their time back. What are they building with it?
