Are You Biased Against AI?

Functional Fixedness

Adopting Generative AI is in many ways a human transformation. This is because we tend to build and stick to mental models of the world and the tools within it to avoid cognitive overload. This is illustrated by the concept of "functional fixedness," famously shown in the 'Candle Problem'.

In the Candle Problem, subjects are given a candle, thumbtacks, and matches and asked to fix the candle to a wall so it won't drip. The solution involves emptying the thumbtack box, tacking it to the wall, and then placing the candle on the box. The difficulty lies in seeing the box not just as a container but as a platform (which requires overcoming the fixed idea of its usual function).

This tendency towards 'functional fixedness' is a barrier to truly leveraging AI. We see an AI tool through the lens of its most obvious, initial functions, like writing text or summarizing, and mentally categorize it as such, much like the subjects who saw the box only as a 'tack container'.

Status Quo & Loss Aversion

This initial, narrow view then feeds a second, more powerful bias: Status Quo Bias. If an AI is perceived as just a slightly better drafting tool, why should one change an entire workflow for what appears to be a marginal gain? The resistance to change feels logical because the perceived benefit is small and contained.

This inertia is compounded by Loss Aversion, where the pain of a perceived loss looms larger than the pleasure of a potential gain. The effort to learn a new process, and the fear of ceding control or devaluing one's existing skillset, feels far more threatening than the incremental benefit of the 'better summarization tool' we believe the AI to be. Consequently, the true, game-changing opportunities of using AI are not considered, because our initial functional fixedness has already devalued the prize.

Reframing

Overcoming these cognitive biases requires a fundamental reframing of the technology. We must move beyond viewing AI as a deterministic tool for retrieving known information, (using it like a knowledge base). Much of its true power lies in its non-deterministic nature as a creative engine, capable of generating what may appear as novel solutions.

The opportunity, therefore, is not just in optimizing existing processes, but in identifying entirely new problems that it can solve when leveraging your unique competencies, data, and networks to create new forms of value.