Artificial Intelligence, the Human Brain’s Randomness, and the Illusion of Superior Decision Making
In 2017, I put forth a theory in both a paper AI article innovation360 and a book (link) that AI’s capacity for effective decision-making has its limits, particularly when we delve into the randomness inherent in the human brain. A few years down the line, this topic remains timely and warrants revisiting.
The Nature of Randomness in the Human Brain
At the heart of this theory is a profound, nuanced understanding of randomness. Unlike the term’s common use, randomness in the context of the human brain isn’t about arbitrary decisions or chaos. Instead, it’s rooted in the unpredictable spiking behavior of neurons. These neurons spike both spontaneously and in response to stimuli. As highlighted by Edmund T. Rolls and Gustavo Deco in their work, “The Noisy Brain: Stochastic Dynamics as a Principle of Brain Function” at Oxford University, this seemingly ‘noisy’ behavior arises from a myriad of factors: genetic variability, electrochemical and thermal disturbances, and even the spiking activity of neighbouring neurons. Reference.
Artificial Intelligence’s Shortcomings
Despite the remarkable strides made in AI, no technology has come close to emulating the intricate, unpredictable nature of the human brain. This doesn’t discount the power of AI. Just as word processors were revolutionary upgrades from typewriters, AI presents monumental advantages for large-scale, structured tasks. Yet, when it comes to decision-making colored by the brain’s unique ‘noise,’ AI remains lacking.
Recent research from Stanford further fuels this conversation. The paper found here indicates a noticeable decline in the performance of ChatGPTs Large Language Models (LLMs) over the past month. It’s a stark reminder that while AI can evolve, it may not necessarily be in the direction we anticipate.
So, What’s the Future for Artificial Intelligence and decision making?
It’s easy to get carried away with AI’s astonishing feats, but one must remember its limitations. Let’s recognize AI for what it is: an immensely powerful tool for specific tasks. But claims of AI taking over nuanced human decision-making remain, at least for now, more speculative than substantial.
As we continue to delve into the future of AI, let’s celebrate the unique aspects of human cognition and consider them in conjunction with AI’s strengths. The interplay of machine precision and human randomness may just be the key to a harmonious and productive future.
