As artificial intelligence becomes an integral part of software development, a fundamental question arises: does AI erode essential engineering skills, or does it pave the way for new capabilities?
This tension is especially pronounced with the growing presence of code assistants and agentic AI: these tools increasingly handle routine coding tasks, raising concerns that traditional programming skills could atrophy.
But perhaps this concern reflects a limited perspective. What if AI isn’t replacing skills, but reshaping them? Why are we not more optimistic about skill improvement through AI?
Part of the issue may lie in how we talk about AI. Unlike other tools—calculators, CAD systems, or test automation frameworks—we often speak of AI in emotional terms, debating whether we “trust” it or “believe” in its capabilities. Popular culture, especially movies, fuels this tendency by portraying AI as an autonomous force that will inevitably harm humanity once it takes over. These narratives reflect a collective bias that subtly inhibits how we adopt AI in professional settings—with caution and sometimes fear.
To move forward, we need to reframe AI not as a rival, but as a tool—one that has its own pros and cons and can extend human capability, not devalue it.
This shift in perspective opens the door to a broader understanding of what it means to be a skilled engineer today. Using AI doesn’t eliminate the need for expertise—it changes the nature of that expertise. Classical programming, once central to the developer’s identity, becomes one part of a larger repertoire. In its place emerge new competencies: critical evaluation, architectural reasoning, prompt literacy, source skepticism, interpretative judgment. These are not hard skills, but meta-cognitive abilities—skills that require us to think about how we think. We’re not losing cognitive effort—we’re relocating it.
This transformation mirrors earlier technological shifts. The calculator didn’t render algebra obsolete—it enabled us to solve more sophisticated problems. CAD tools didn’t eliminate design—they replaced manual drafting with new creative possibilities. In each case, the locus of value moved from mechanical execution to higher-order thinking. AI is pushing us along a similar trajectory.
Yet despite this evolution, many organizations remain anchored to outdated metrics. Developers are still assessed primarily on their ability to produce code by hand, rather than on their effectiveness at leveraging AI tools to improve outcomes. It’s akin to evaluating a loom operator by how well they stitch by hand. The value has shifted—from manual dexterity to system-level thinking. Modern software development now requires skills in articulating intent, refining outputs, and integrating automated suggestions into coherent products.
Still, most businesses lag behind. While many executives extol AI’s potential, they quietly shift the burden of adaptation onto employees. Reskilling is rarely structured or funded; it’s expected that workers upskill on their own or risk becoming obsolete. As Ford CEO Jim Farley bluntly predicted, “AI is going to replace literally half of all white-collar workers in the U.S.” The middle tier—too senior for retraining bootcamps but not steeped in emerging tools—finds itself squeezed out not by algorithms, but by inaction from leadership. This approach raises stress levels for employees, leading either to burnout as they try to handle everything themselves, or to anxiety as they struggle to find their place in the new reality—both of which ultimately result in decreased productivity.
Yet this trajectory isn’t inevitable. Companies like Accenture have committed to large-scale interventions—investing $3 billion to double their AI talent to 80,000 through hiring, acquisitions, and internal training. Others, like Microsoft and TD Bank, are embedding AI fluency into team structures and performance metrics. A recent survey of GitHub Copilot users at TD found 75% felt equally or more productive, while firms like Tapestry and Levi’s report measurable efficiency gains. Importantly, these organizations aren’t just reducing headcount—they’re redefining roles and retraining talent to operate at a higher level of abstraction.
Some of the early adopters of AI enablement are already looking ahead—not just at the savings from replacing employees with AI, but at the additional gains those savings might unlock. With strategic investment and redesigned expectations, AI can become a growth driver—not just a cost-cutting tool.
But upskilling alone isn’t enough. As organizations embed AI deeper into the development workflow, they must also confront the technical risks that come with automation. The promise of increased productivity can be undermined if these tools are applied without adequate context, oversight, or infrastructure.
AI-generated code can introduce maintainability issues, hallucinations, and security vulnerabilities—especially when used passively or without context. But these are solvable problems. The path forward lies in building engineering environments with robust feedback loops, automated compliance checks, and quality enforcement mechanisms tailored to each domain. Teams must also establish architectural and ethical “guardrails” that guide both humans and machines toward better outputs. That also means transforming the development skillset toward built-in quality thinking—designing and reasoning before generation, rather than relying on “it will be tested later” after the code has already been produced.
In addition to technical and organizational dimensions, this transformation signals a deeper philosophical shift. Some engineers may argue that with AI, being reskilled, they risk becoming not creators but merely output reviewers. But there’s no need to choose between creation and criticism. In The Critic as Artist, Oscar Wilde challenges the notion that these are distinct roles. He elevates the critic—not as a passive evaluator, but as a creative force who imposes structure, interprets meaning, and gives form to complexity. His vision feels increasingly relevant in the age of AI development. As machines take on the mechanical aspects of software construction, developers are stepping into a more curatorial role. Their value lies in how they interpret, adapt, and orchestrate—not merely how they build. Engineering, like art, is becoming less about the brushstroke and more about the composition.
We are not simply building with new tools—we are redefining what it means to build. To unlock the full potential of AI, organizations must rethink how they measure contribution, invest in reskilling, and embrace a broader definition of engineering excellence.
