03/06/2026
Engineers are increasingly using AI tools to draft sections of engineering documents, design basis sections, equipment philosophies, specification clauses.
And the output is often impressive looking. Structurally correct. Formatted properly and Confident in tone.
But here is what a software instructor observed about his students using AI to solve engineering challenges, and it applies directly to what is happening in process engineering right now.
"The AI code always worked. But it was consistently overcomplicated, harder to maintain and harder to scale than what I taught. Every time, my solution was cleaner. Because the AI was being limited by the user who did not know enough to prompt it well or to know when it was solving the wrong problem efficiently."
Replace code with engineering philosophy and you have an exact description of what happens when a process engineer uses AI to draft a design basis document, they do not fully understand.
The AI will produce a sparing philosophy section. But if the engineer does not understand the difference between a criticality-based sparing decision and a cost-optimized one, they cannot tell whether the AI chose the right basis for their specific application.
The AI will draft a pressure relief philosophy. But if the engineer does not understand the IBR requirements and corrosion allowance implications that should be shaping that philosophy, they cannot catch what the AI got subtly wrong.
A design basis document signed off by an engineer who does not understand its contents is not a document. It is a liability waiting to surface at HAZOP, at commissioning, or worse.
Our Process Engineering Design Basis course covers all 19 sections of a Design Basis Document with a focus on the engineering rationale behind each decision, not just what to write, but why.