SWPS University

Mateusz Czopek is Head of the Center for Online Learning at SWPS University and brings 15 years of experience across academic and corporate e-learning environments. His work focuses on agile production, instructional quality, accessibility, and the thoughtful use of AI to support course design, localization, evaluation, and scalable digital learning delivery.

With 15 years of experience across corporate and academic environments, what key lessons have shaped your approach to e-learning technical production? 

One of the most significant lessons I have learned is that learning needs and goals, not technology, must drive production. Whether designing for a university or a multinational corporation, the format of the material should be dictated by the subject matter, the target audience, and the intended learning outcomes. Early in my career, e-learning was often about "the next big thing"—flashy animations and complex interactivities were the markers of quality. Today, we live in an era of digital fatigue and information overload. Modern learners are overstimulated. They don't need more "bells and whistles."

They need clarity, methodical design, and high-value content. The focus has shifted from how much tech we can pack into a course to how effectively we can design an educational experience that respects the learner’s time and cognitive load while addressing their real learning needs. Speaking of needs, we should also consider people with special needs when designing and developing online courses. Creating fully accessible digital educational materials has always been a significant challenge for creators, mainly due to the increased time required for production. Fortunately, this process is now being streamlined, driven by technological advancements, particularly in the field of AI. 

How is AI influencing the design, development, and delivery of online learning today? 

It probably won’t surprise you to hear that the biggest immediate impact of AI on online learning is efficiency. When used skillfully, AI may considerably reduce production costs and speed up the design and development of learning materials. At our Center, we use AI across all these stages, though it shines brightest in technical production. Just to give you an idea, we leverage generative AI for creating graphics, infographics, and professional voiceovers. We also use it to dub and translate videos for course localization.

On top of that, AI has revolutionized our course evaluation process. By using it to group and analyze thousands of student comments each semester, we have turned a slow, manual task into a fast, data-driven workflow. However, we are really cautious when it comes to using AI for actual instructional design and content development. Based on our experience and observations, human expertise is still essential for creating meaningful learning paths and developing high-quality content. We see AI as a supporting tool, not as the architect.

Because AI can sometimes "hallucinate" or make things up, a human should always be involved to verify the accuracy and quality of the final product. Looking ahead, I am excited about the potential for hyper-personalized learning and AI assistants, but I believe that human oversight will always remain the most important part of the process. 

What does agile e-learning production mean in practice? 

For me, agile e-learning production means shifting from linear, "all-at-once" development to an iterative, value-driven process. Instead of working in isolation to finish a complete course before testing, we deliver functional "learning increments" in short, repeatable cycles. This allows continuous feedback and rapid adjustments, ensuring the final output is always aligned with the stakeholders’ expectations. In practice, this usually means prioritizing prototyping over exhaustive documentation, establishing tight feedback loops, and fostering genuine cross-functional collaboration. When instructional designers, subject matter experts, and developers operate as a unified team rather than in silos, the project benefits from shared ownership and collective problem-solving. 

How can institutions implement it effectively? 

Implementing agile, including agile e-learning production, is a transformative process. I believe there is no ‘‘one-size-fits-all’’ approach; rather, it requires a strategic shift in organizational culture. To implement this effectively, I would focus on three core pillars. First, alignment and education. Everyone involved must share the same definition of what "agile" means in their specific context. This requires hands-on training to ensure people fully understand the mechanics of the new workflow. Second, mindset transformation.

The biggest challenges are often cultural. We are asking people to move away from comfortable, traditional habits, which requires clear communication and patience. Those leading the agile implementation process must actively drive this change by fostering a transparent and collaborative environment. In my view, the third pillar is about leading with value. Change is easier to embrace when the benefits are clearly visible. I always use a "language of value" when speaking to stakeholders. Instead of talking about process jargon, I focus on results. For example: "By showing you progress every two weeks, we can fix issues early and ensure the final product is exactly what you and your learners need." By constantly highlighting these tangible advantages—such as reduced rework and higher quality—we are sometimes able to turn skeptics into advocates. Ultimately, successful agile e-learning production implementation isn't just about changing how we work. It’s about proving that this approach delivers superior educational outcomes, and if we ignore even one of these pillars, the whole process may not work. 

As technology accelerates, what major challenges are redefining quality and scalability in online education? 

From my perspective, the greatest challenge for learners today is the proper selection of high-quality educational materials. As GenAI tools become more accessible, I am concerned about the market being flooded with low-quality, unverified educational content—a trend we are already beginning to see. AI works best for those who are already experts. It amplifies their existing efficiency.

However, for those without deep domain knowledge, AI can be a trap. If you lack expertise in instructional design, you cannot verify whether AI-generated output, such as a training curriculum, is accurate or pedagogically sound. While the growing demand for 'lifelong learning' is a positive sign of a proactive society, the ease of using GenAI might tempt many to bypass the necessary instructional and technical rigor. The result? Cheaper courses that look professional but lack substance. Ultimately, the challenge of the future is not creating content—it is curating and verifying its quality. 

What advice would you offer to emerging e-learning professionals navigating technical production in the age of AI? 

My advice is simple: invest in yourself before your toolkit. Become a true expert in instructional design and e-learning development. Only when you possess a deep understanding of how people learn can you effectively direct AI tools to support your work. Without that foundational expertise, you risk contributing to the wave of low-quality content I mentioned earlier. Use AI to augment your skills, but never let it replace your critical thinking.