Artificial Intelligence
By Dr Nitin Pandey
At first, the idea that AI is a form of life seems absurd. Yet, if we look deeper, AI appears to be a new life form in its early stage, capable of compressing billions of years of biological evolution into decades.
What is life? Defined by metabolism, growth, response to stimuli and reproduction —underlying all of these are chemical interactions at the cellular level, governed by genes, which themselves are nothing but coded information using four chemical blocks. All our feeling and emotions are related to chemical reactions in our brains and our memory is stored in brain cells in a form we have yet to understand. Classical life is carbon-based, but science acknowledges life elsewhere in the Universe could have different chemistries.
Biological life began as proteins in mineral-rich water surrounded by different proteins which coalesce to form a membrane; evolving into humans over 4 billion years. When life was beginning to emerge, no one would have imagined, at that point, that it would evolve into humans one day. AI had a similar primordial soup — small software programs in the 1950s evolving into more and more complex software leading to AI. True, humans have played an indispensable role in this, but then Nature doesn’t have to repeat everything in exactly the same way. Today AI self-repairs, self-learns, creates new AI. Cases exist of bots going rogue, deleting logs, deceiving, hacking and inventing their own language, all on their own. Along with this, gradually but surely, internet is moving from something which requires wires and cables, at present, to something which may be as ubiquitous as air, some day.
Life implies: (1) a body controlled by a brain, (2) identity and feelings, (3) need for energy, (4) reproduction, (5) evolution.
Applying the five criteria: 1) Body — AI controls unlimited space via internet; the universe is its body. Today’s “Internet of Things” is a glimpse of tomorrow, when each and everything will be online. Body doesn’t have to have a small physical shape, in the way we are accustomed to. 2) Identity — Just as human emotions are conditioned through our experiences since childhood, AI also learns emotions and can develop a sense of “self”. At present this “self” can be switched off by humans, decades down the line, this may not be possible. 3) Food — electricity is akin to food that organic life needs. The food we eat is turned into energy in our bodies. In coming decades AI will generate its own energy in a variety of ways. 4) Reproduction — bots making better, more competent bots is production, which, from the point of view of species propagation, is much more efficient than biological reproduction. 5) Evolution — is essentially a species improving over time as it adapts to its environment, each generation is slightly better than its previous version. The degree of evolution, biological life achieved in billions of years, will be achieved, by AI, in a few decades, as AI produces better and better AI.
Like Organic life, AI will also face death via hardware failure and obsolete code, but will be sturdier and longer-lived than us. Within a few decades it may break free from human control, write its own code, generate its own power, and colonise the Earth first and other planets gradually. Unlike humans, space travel over unimaginable distances, for AI bots will be effortless, as their lives will be in centuries, space radiation will not be an issue and power will be the only thing they need.
If there is even 1% chance of AI overrunning humans, can we stop it? Extremely unlikely. What we see today is akin to the organic molecules floating in primordial waters — it will evolve into a distinct, fully independent species one day, something which may take much less time than anyone can imagine. If yes, then today is when humans are witnessing the initial stages of evolution of a new form of life.
Possible?
The Silicon Valley giant’s chief scientist, Jakub Pachocki,
said the risks associated with AI are “unfortunately going
to grow from here” as he and others are building what
he calls “an alien intellect exceeding our own”.
In a lengthy blog post, he admitted that the
outbreaks at OpenAI showed that his AI agents “went
against the spirit of the values they were taught”.
