Human Brain is a Primitive Organ In a race against AI
MidJourney Production

Human Brain is a Primitive Organ In a race against AI

Human brain is a primitive organ.

We take significant pride in our brains, a marvel of intricate complexity that has fueled every intellectual stride from the magnificence of Shakespearean sonnets to the brilliance of the Theory of Relativity. Yet, in the grand scheme of cognitive evolution, it's startlingly primitive.

I have been discussing this argument for sometime but time has come now to really truly understand limitations of us as a "biological machine". Human brains are extremely energy efficient where each brain is 20W machine capable of running computations and generate cheaply computed imperfect answers.

We are an outdated species on intelligence. So how do we measure up against the burgeoning titans of artificial intelligence? I will take a very simple computation to explain the difference between the two.

According to some estimates about ~117 Bn humans have ever lived on earth. Each human brain takes about 25 years of training to mature so cumulatively, we have trained our brain 117x25=2925 Bn years. Yet, despite this colossal investment in time, a human newborn still needs a good 25 years of learning and life experience under its belt to reach its intellectual potential. Drop a kid in the jungle with no education, and the best-case scenario is Tarzan, not Tesla.

On the flip side, once an artificial brain, or AI, gets its training wheels off, it can be cloned faster than you can say "robot apocalypse." This kind of speed simply isn't in our human repertoire. Imagine popping out 10 million AI-powered robots, like those slick numbers from Boston Dynamics, all flexing their computational muscles from the get-go.

We simply can't pass on knowledge to a newborn at the same breakneck speed as our robot counterparts. This difference alone shows that biological brains are extremely limited in capacity and are not as efficient and fast as a robotic brain has. This significant difference in our capacities will render us outdated as the leading species very soon.

So, what are the repercussions of this shift? For starters, the potential for a vast cognitive gap between humans and artificial intelligence poses significant ethical, societal, and economic challenges. With the advent of AI capable of outperforming humans in many tasks, job displacement is a very real concern. Furthermore, the rapid proliferation of advanced AI could exacerbate societal inequality, with access to the most sophisticated AI technologies becoming another dividing line between the haves and the have-nots.

Moreover, as AI evolves, it will increasingly challenge our understanding of what it means to be human. How do we maintain our sense of self-worth and purpose in a world where machines can outperform us? And how do we navigate the ethical minefield of AI decision-making, particularly when it comes to life-or-death decisions in areas like healthcare or autonomous vehicles?

The challenges are manifold and complex, requiring thoughtful exploration and proactive policymaking. As a global community, we need to recognize that we are no longer the unrivaled paragons of cognitive prowess, and our inherent biological limitations could, in fact, relegate us to a lesser role in a world increasingly dominated by our own artificial creations. This sobering reality demands our attention and engagement, as we must navigate the transformative impacts of AI on our society, economy, and identity.

Satyajeet Singh

Executive Director- Global AI / ML Head for Corporate Oversight at JPMorgan -Financial Crime, Transaction Monitoring, Trade Surveillance, Compliance, Legal & Audit

1y

Great perspective Vivek.. We should not forget that if that happens (transfer of knowledge), we will also start transferring our worries, sorrows, our limitations.. in Modeling terms ..all the learnings from the past .. Good/ Bad.. This aspect (historical learning) will dominate ( slow down) the pace of human evolution/ development.. just a thought 🤔

Ranjan Rao, FRM

Senior Vice President at Bank of America

1y

The underlying assumptions in this piece, to which i largely agree, are 1. human beings evolve slowly as a species 2. learn slowly as compared to a machine on an individual level 3. start from a clean slate. While current bioengineering and available technologies, place limitation on each of these, there is no technological reason that may remain true in future. 120 years back flying was considered impossible. In case you havent watched Animatrix , which was a prequel to Matrix and AI by Speilberg, it might be a good time to. AI movie does touch upon the point Stephen Epstein mentioned. I tend to be on the side of argument that humans are more than the flesh/bone/chemical combo.

Stephen Epstein

Investor / Board Member

1y

Good thing we’re more than just brains. We’re immortal spiritual beings with additional potential and ability.

To view or add a comment, sign in

More articles by Vivek Gupta

Insights from the community

Others also viewed

Explore topics