It became clear that code alone was insufficient at one point, somewhere between the development of facial recognition software and the first congressional hearing regarding teenage social media use. Even though engineers could create incredibly complex systems, they consistently produced results that confused, alienated, or subtly hurt the people who used them. There was something lacking. As it happened, it was an awareness of people.
For many years, behavioral psychology and computer science read different journals, lived in different buildings, and hardly ever interacted. Logic problems were resolved by computer scientists. Psychologists investigated the reasons behind people’s illogical choices. The overlap appeared insignificant, almost coincidental. However, observing how this has changed over the past few years, it’s difficult to ignore the fact that the two fields have practically merged due to necessity rather than any official agreement.
The current state of affairs was revealed at the 2024 Consumer Electronics Show in Las Vegas. Twenty years ago, the American Psychological Association and the Consumer Technology Association would have had very little to talk about, but now they shared a stage. On panels discussing AI ethics, privacy design, and digital therapies, psychologists spoke with the same authority that was previously reserved for software architects. That’s a big change. Who gets to decide what technology becomes has changed structurally.
This merger was prompted in part by mounting evidence that AI systems lacking psychological foundation actually cause harm. Early facial recognition software produced error rates that resulted in erroneous arrests because it was trained almost exclusively on lighter skin tones. These weren’t conventional software bugs. These were mistakes in human comprehension—failures to inquire about whose experience the system was truly intended to support during the design phase. The individuals creating these potent systems make up a remarkably small portion of the global population, according to Nathaniel Fast, director of an ethics center at USC’s Marshall School.
Additionally, there is the issue of trust, which has a profoundly psychological component. It is possible to create an AI system that is technically perfect but that people won’t use because they perceive it as dangerous or opaque. According to David Luxton, a clinical psychologist at the University of Washington, trust in AI functions on two different levels: trust in the machine itself and trust in the people who operate it. Building and maintaining both requires psychological expertise. Better code doesn’t solve either.
Another area where the two disciplines have clashed in ways that seem more urgent is privacy. Current privacy controls, such as the cascading permission screens that most people tap through without reading, are just too cognitively taxing for average users, according to Heng Xu of the University of Florida. Understanding how people actually process information under pressure is necessary to design something that is truly usable; this is a psychological issue before it becomes a design issue.
Whether academic institutions are progressing quickly enough to formalize this partnership is still up for debate. Although there are interdisciplinary programs, they are still dispersed and have uneven funding. The institutional frameworks that support the research are not keeping up with its rapid advancement. In the meantime, AI labs, gaming studios, and health tech firms are hiring behavioral scientists alongside engineers without much fanfare, acting as if the merger has already been finalized.
Observing all of this, there’s a sense that what initially appeared to be an unlikely pairing has subtly evolved into the industry’s standard operating procedure. Technology was not replaced by psychology. Technology came to the somewhat delayed realization that it had always required psychology to make sense of itself.
