Exploring the technical, ethical, and practical dimensions of AI through the lens of Mechanical Engineering and human choice.
Explore FoundationsGenerative Pre-trained Transformers (GPT) are not knowledge engines; they are statistical prediction models. They predict the next most likely token in a sequence based on vast training data.
For an engineer, this means recognizing that AI hallucinations are simply statistically likely but physically impossible predictions.
While discriminative AI classifies data (e.g., Identifying stress fractures in a CAD model), generative AI produces new data (e.g., Drafting the executive summary of a laboratory report).
True professional literacy requires more than just knowing how to prompt. It requires a comprehensive understanding of the sociotechnical ecosystem.
Understanding model architecture, training data limits, and the statistical nature of outputs.
Competence in using tools to generate engineering artifacts while maintaining data integrity.
Responsibility for bias, data privacy, environmental impact, and professional stewardship.
Questioning the broader impacts of automation on the engineering profession and society.
Framework by Tadimalla, Cary, Hull, Register, Heafner, Maxwell, and Pugalee (2025)
As a mechanical engineer, AI is a collaborator, not a replacement. Human agency means you decide when to use a tool, how to verify its output, and when to reject its suggestions entirely based on physical laws and ethical obligations.
We use Generative AI as a scaffold—a temporary support that helps you learn the structure of technical memos, professional verbs, and report conventions.
Professional Tip: Use AI to suggest an outline for your lab report based on your raw data. Then, populate the outline yourself to ensure the engineering analysis is sound.