Are Any Engineering Jobs Safe From AI?
This is a fair question for students to ask before committing four (or more) years and a lot of money to engineering degrees. It’s also a question that executives, hiring managers and frontline engineers are actively debating right now. Here’s a snapshot of the current conversation.
The fastest-growing jobs are the ones closest to AI and energy technologies
The World Economic Forum’s 2025 Future of Jobs report found that AI-related and technical roles are among the fastest-growing job categories in percentage terms. Roles tied to the energy transition, including environmental engineers, renewable energy engineers and electric and autonomous vehicle specialists, are also among the fastest-growing professions in the report.
Construction leaders see AI as a co-pilot, not a replacement, for engineers
Autodesk recently surveyed more than two dozen construction executives and engineers about what they expect from AI in the year ahead. Although Autodesk does have a bias here — they sell AI tools to the construction industry — the on-the-record predictions from executives are still informative. One executive called AI “a professional enabler, not a substitute for expertise,” arguing that the technology can help human engineers and architects solve more complex challenges than they could on their own. Another executive made a point that most AI systems are trained on documents and schedules, not the physical realities of a job site. Although AI can create an optimized plan, only a human engineer physically present on site can catch problems the model won’t see coming.
Disruption in software engineering is concentrated at the entry level
A Harvard study tracking 62 million workers and 285,000 firms found that, after firms began adopting generative AI in early 2023, junior employment fell 7-12% while senior employment continued to grow. In a 2025 CIO article, Rachit Gupta, head of AI at Tredence, traced that development to the fact that skills junior engineers used to learn on the job, like fixing bugs and writing boilerplate, “are now the kinds of things AI tools handle well.” Raymond Kok, CEO of the development platform Mendix, predicted that developers will shift "from being coders to being what I call composers." Chirag Agrawal, a senior software engineer at Amazon, elaborated on that distinction: “My role has shifted from just coding to validating AI output, checking for edge cases, security risks and logic gaps that AI can’t catch.”
What this means for today’s engineering students
No engineering discipline is definitively “safer” than another. It has become clear that the kinds of work most exposed to AI right now are the routine, well-specified tasks common to entry-level positions. Every engineering specialty has some of that work, alongside the judgment-heavy work that none of these sources expect AI to take over soon. The practical takeaway is to treat fluency with AI tools as part of the engineering degree itself, since it’s become clear that such fluency is the dividing line between who gets an entry-level position and who doesn’t.
