The rapid development and widespread adoption of Artificial Intelligence (AI) technologies have made the development of AI-specific design methods an important topic to advance. In recent decades, the centre of gravity in AI has shifted away from cognitive science and related fields like psychology. However, there is a clear need and potential for added value in returning to stronger interaction. One potential challenge for this interaction may be the lack of common conceptual grounds and design languages.

In this article, we aim to contribute to the development of conceptual interfaces for human-based AI-specific design methods through the idea of cognitive mimetics. We begin by introducing basic concepts from mimetic design and interpret them in the context of this thematic area. These provide some of the basic building blocks for a design language and bring to the surface key questions. These in turn provide a ground for explicating cognitive mimetics. In the second part of this paper, we focus on specifying a key aspect in cognitive mimetics: the contents of information processes.

Others engaged in this field can derive value from using or developing the basic conceptual machinery to specify their own approaches in this interdisciplinary field that is still shaping itself. Furthermore, those who resonate with the idea of cognitive mimetics, as specified here, can join in taking this particular approach further.