Lecture on the use of generative artificial intelligence (GenAI) in chemical process design.

Generative artificial intelligence (GenAI) is rapidly emerging as a powerful new tool in chemical engineering, with the potential to fundamentally change how chemical processes are designed, documented, and evaluated. Unlike traditional machine-learning approaches focusing on prediction or optimization within predefined structures, GenAI models can generate new content—such as process flow diagrams, equipment specifications, design alternatives, and documentation—directly from learned patterns in historical data and engineering knowledge bases.

This talk provides an accessible introduction to generative AI for chemical process design, distinguishing it from other AI and machine-learning methodologies commonly used in the industry. Key applications are discussed, including automated design assistance, digitization of PFDs and P&IDs, process documentation, and early-stage HAZOP support. The role of large language models, fine-tuning, and retrieval-augmented generation in enabling domain-specific engineering workflows is highlighted through practical examples.

In addition to opportunities, the talk addresses critical challenges that must be overcome for responsible industrial adoption, including model transparency, validation in safety-critical environments, integration with legacy engineering tools, workforce upskilling, and governance. The session concludes with a perspective on where generative AI is most likely to deliver near-term value in process design and how chemical engineers can engage with these tools safely and effectively as the technology matures.