Why AI Isn't Ready for Commoditization

Technology Life-cycle

As I observe the current state of Artificial Intelligence (AI) and the rush surrounding its deployment, I find myself reflecting on a pattern that has repeated throughout technological history—a life-cycle we should follow or ignore at our peril. Understanding this cycle will be crucial as we navigate the turbulent waters of machine intelligence in the coming decades.

Technology Birth: The Age of Polymaths

At the start of something new, technology emerges from the minds of individuals who must be both theorists and builders out of necessity. During this nascent phase, technology represents the promise of future benefits—a tantalizing glimpse of what could be possible if we can unlock nature's secrets. But here's the thing: these pioneers cannot simply theorize; they must also engineer the very methods and means to test their theories and conduct their experiments.

I think of figures like Alan Turing, who didn't just conceive of computation as a mathematical abstraction but had to grapple with the practical challenges of building machines that could embody his ideas. Robert Oppenheimer, who couldn't rely on existing infrastructure but had to orchestrate the creation of entirely new engineering capabilities to transform theoretical physics into reality. Niels Bohr, whose quantum insights required him to work hand-in-hand with experimentalists and instrument makers to probe the atomic realm.

These pioneers are remembered not as narrow specialists, but as polymaths who had no choice but to embody both scientific curiosity and engineering necessity in a single person. They were forced to be polymaths because the specialized infrastructure we take for granted today simply didn't exist. They had to build their own tools, design their own experiments, and create their own methods for testing the boundaries of the possible.

At this stage, the technology exists primarily in the realm of possibility, but that possibility can only be explored through ingenious combinations of theory and practice. The science dominates the vision, but the engineering dominates the day-to-day reality of actually making progress. We explore uncharted territory where both the map and the vehicle must be invented simultaneously.

Technology Maturation: The Great Separation

This pioneering phase, however, cannot sustain itself indefinitely. As we look at the evolution of any transformative technology, science and engineering eventually must part ways to serve the technology's evolution. This separation marks the beginning of true maturation—when technology transitions from promise to realizing that promise.

During this critical phase, we see the emergence of engineering as a distinct discipline with its own methodologies, constraints, and objectives. While scientists continue to push the boundaries of what's theoretically possible, engineers focus on the art of the practical: How do we make this work reliably? How do we scale it? How do we manage its complexity and cost?

This separation isn't arbitrary—it's a natural evolution that allows each discipline to flourish. This is where engineering truly comes into its own. The theoretical insights gained during the science-dominated birth phase become the raw materials for solving real-world problems. We see the development of standardized practices, specialized tools, and systematic approaches to implementation. The technology gains structure, reliability, and predictability.

Technology Industrialization: The Commodity Phase

The maturation phase gradually gives way to something entirely different. As we look at the next phase of the technology life-cycle, mature technologies enter their final phase: widespread adoption through scaling and refinement. At this stage, technology becomes a utility and commodity, much like electricity or telecommunications today. The focus shifts from fundamental innovation to assembly, component refinement, and optimization.

This transformation has its purpose. The cutting-edge science becomes background knowledge. The specialized engineering practices become standardized procedures. The technology that once required polymaths, scientists & engineers, now operates through well-understood processes and established infrastructure.

This is precisely where I believe Information Technology finds itself now. The days of inventing new information technology paradigms have largely passed. Instead, we are in an era of integration, standardization, and incremental improvement. Agile is a perfect example of this, as we care less about engineering the technology stack rather than using it. The science is well-established, the engineering principles are codified, and the primary challenge becomes efficient deployment at scale.

History Repeating

As I look at the current state of artificial intelligence, I see clear parallels to this historical pattern. We are witnessing the emergence of our modern equivalents of Bohr, Oppenheimer, and Turing—visionaries who are simultaneously advancing the science of intelligence while grappling with its practical implications. The field remains dominated by scientific discovery, with engineering practices still in their infancy.

However, I am already seeing early signs of the great separation beginning. As AI moves beyond pure research, distinct engineering domains are starting to crystallize. We are beginning to see the emergence of specialized practices around model deployment, safety engineering, human-AI interaction design, and scalable training infrastructure.

This mirrors exactly what happened with previous transformative technologies. The science-engineering split is starting to happen, though many haven't recognized it yet.

The Critical Mistake We Must Avoid

Here is where I believe we are making a fundamental error. Too many organizations and leaders are treating AI as if it were already in the commodity phase—ready for immediate, large-scale adoption with minimal specialized expertise. This represents a dangerous misunderstanding of where we actually stand in the technology life-cycle.

This misconception has real consequences. AI should not be rushed into the utility and commodity stage while skipping the crucial engineering maturation phase. Just as we wouldn't have expected the early pioneers of computing to immediately build data centres, we shouldn't expect AI to seamlessly integrate into every business process without first developing robust engineering practices.

The consequences of this premature commoditization are already becoming apparent. We see systems deployed without adequate safety measures, unrealistic expectations about reliability and performance, and a general underestimation of the specialized knowledge required to implement AI effectively.

Purpose in the Process

As I think about the path ahead, I am convinced that respecting this technological life-cycle will be essential for realizing AI's full potential. We must allow the engineering phase to unfold naturally, developing the specialized practices and institutional knowledge necessary for responsible deployment.

This requires a fundamental shift in expectations. This means accepting that we are still in the early stages of a much longer journey. The scientists continue their essential work of expanding the boundaries of what's possible, while a new generation of AI engineers is already emerging to bridge the gap between laboratory breakthroughs and real-world applications.

The technology life-cycle teaches us that shortcuts are illusions. Each phase serves a purpose, and attempting to bypass any stage risks undermining the entire enterprise. As we stand at this critical juncture in the development of artificial intelligence, I believe our patience and respect for this natural progression will determine whether AI becomes a transformative force for good or another cautionary tale of technological hubris.

The future of AI—and perhaps the future of human progress itself—depends on our wisdom to let this life-cycle unfold as it should, rather than as we wish it would.

---

About the Author:

Raimund Laqua, P.Eng, is a professional computer engineer with over 30 years of expertise in high-risk and regulated industries, specializing in lean methodologies and operational compliance. He is the founder of Lean Compliance and co-founder of ProfessionalEngineers.AI, organizations dedicated to advancing engineering excellence.

As a Professional Digital/AI Engineering Advocate, Raimund champions proper licensure across the entire spectrum of digital engineering disciplines. He actively contributes to the profession through his leadership roles, serving as AI Committee Chair for Engineers for the Profession (E4P) and as a member of the Ontario Society of Professional Engineers (OSPE) working group on AI in Engineering, where he helps shape the future of professional engineering practice in the digital domain.