Don’t Worry About The AI Singularity: The Tipping Point Is Already Here
As the AI market expands and AI use cases permeate every industry, every once in a while I hear the question – when will the AI singularity occur? For those who are not familiar with this term – the AI singularity refers to an event where the AIs in our lives either become self aware, or reach an ability for continuous improvement so powerful that it will evolve beyond our control. While this is a reasonable concern in the future, I argue that there are much more pressing concerns in the present – in particular that AI has reached a Tipping Point.
A tipping point is a state where a technology grows and permeates our lives very rapidly, building upon itself. The distinction between the singularity and the tipping point, in my view, is that the tipping point focuses on permeation, not intelligence. The AIs that we deal with today are not particularly smart when compared to the human brain. However, from the time we wake up in the morning to when we go to bed, they are everywhere, from the alarm that wakes us to the route we take to work (pre-pandemic!) to countless decisions made behind the scenes by corporations and governments that affect what loan interest rate we get, how we qualify for assistance, decisions on our health, and more. This is the tipping point I am referring to, and while it is far less entertaining than contemplating an all knowing AI overlord, it is real, it is here, and it is affecting all of our lives.
Some examples of the permeation of AI for everyday humans, in both positive and negative ways:
So, what does this all mean? To appreciate that, I looked back at another technology that now permeates our lives – cars.
The figure below shows a 1909 B-Type Racer, the fastest car of its time. To drive this car, the driver needed to explicitly control the air/fuel mixture and the point where the spark plug met the ignition. The driver was not just there to direct the car but was also the caretaker of the engine.
MORE FOR YOU
Fast forward to today. Most of us have either driven a car or been in one, but have no idea what the air/fuel mixture is or how the engine works. The use of cars has permeated, and relatively safely. What has enabled this process? First we have licenses, where we are required to learn and become literate about driving, road rules, etc. before we are allowed to drive a car. The cars themselves are evolving constantly and our interface is becoming higher and higher level (self driving cars are only looking for direction on where to go).
While cars started in the sole domain of mechanical and automotive engineers, now safe cars are everyone’s concern. To me, the key thing that enabled a relatively safe tipping point for cars was the education of the broad public. We know what we need in order to make broad and safe use of the new technology.
As AI permeates, I would argue we need a similar level of education – AI Literacy, since AI much like cars may have started out as a computer science technology but is now directly in reach of our lives and our health.
Concepts: What is AI? How does it work? What are its strengths and limitations? This is not at a mathematical level or a computer science PhD level, but rather at a practical level.
Context: How is AI used? What devices and services around me use AI and how do they use my information?
Capability: Am I able to make decisions in my life based on my understanding of AI Concepts and Context? This may include applying AI technology, or it may mean using and navigating these technologies.
Creativity: As more people become AI literate, we can expect a greater creativity in not just how these technologies are created but also a more inclusive approach to creating and using AI in a way that benefits all.
In conclusion, long before we reach any AI singularity, I expect our immediate challenges will be with the sheer immersion of AI in our daily lives. How we all learn about AI, and become AI literate, will in turn affect the development of the technology and any coming singularity, if it so occurs.