Beyond AI: Design Automation Futures

The explosion of Artificial Intelligence (AI) in the form of Large Language Models (LLM) such as chatGPT and others has revolutionised many aspects of engineering and design, with particular benefits in areas of low level design tasks including code generation, hardware description language model generation, development of test benches, and data analytics. The results are generally good, however still require significant validation to ensure both adequate quality and repeatability prior to deployment, but there is a clear benefit in terms of productivity for engineers in many scenarios. There is however even greater potential for much wider and deeper use of AI methods where greater creativity and exploration could be envisaged including novel topologies, conceptual exploration, creative design, morphic and adaptive design, design space exploration and the identification of novel design features from empirical, simulation and analytical data sources. This paper will explore the further potential of AI and identify approaches that can be applied in design automation to achieve greater leverage of the power of AI in practical scenarios.