Why AI Changes the HCI and UX Practices

The human-computer interaction (HCI) practice of research and user experience (UX) design largely does not exist without the computer. Computer in a superficial sense can mean any programmed tool that performs computations - for example, an abacus could be considered an early practical form of an analog computer.

What today's generations are most familiar with are digital computers — you are looking at one to read this webpage. When computers were made relevant in the post-WW2 era, they took up an entire room and required senior experts who were physically capable of moving mechanisms for the computer to complete even a simple operation. We owe the modern computer to the oN-Line System (NLS), which paved the way for innovators through the 1970s - 2000s to develop and iterate on the Graphical User Interface (GUI).

Through the past 3 decades, our understanding of the interaction with a UI have felt pretty consistent — users interact with a mouse, keyboard, and/or gestures to run software programs on a computer, input values, and extract output from the computer. We have grounded research within the variances of these interactions, we have designed software to evolve and meet users' needs, and we have reached across disciplines to enrich our understanding of computers and their impact on our behavior, our actions, and our culture.

In November 2022, OpenAI publicly released ChatGPT. The webapp itself was positioned as a research release to chat with a large language model (LLM), and while prone to errors, users flocked to the site to break-test countless use cases without end. The past two and a half years have been nothing short of competition, advancements, and geopolitical war to push AI as the flagship interface for the future of computing.

HCI professionals now have a conundrum to work with — humans have now harmonized with artificial intelligence (AI), leveraging it to be more efficient with work, create new content, learn about complex topics...some even going as far to have full-context relationships with these AI models.

How do we define and facilitate this harmonization between man and machine?

This is the burning question that has shifted HCI researchers to explore empirical understandings of AI contexts; it is also a question that has motivated UX designers to create new and interactive experiences to learn of the value of AI in practical applications. At this crucial inflection point, these two practices will need to leverage the other's respective findings and innovations to advance AI systems that humans can make use of.