Unlike humans, computers scale like nothing before – Carl Bass on TEDx Berkley.
This quote is from a TEDx talk on the ‘New Rules of Innovation’ by Carl Bass, the president and CEO of AutoDesk Inc. It was quite an explorative session on innovation. He had spoken about a variety of topics around the world of innovation. One among his themes was the coinage ‘Infinite computing’. Infinite computing? the terms sounded interesting and weird to me at the same time ! I just googled by impulse and curiosity.
One snippet from Wikipedia says
A quantum mechanical system which somehow uses an infinite superposition of states to compute a non-computable function.
Thus my preliminary analysis suggested to me that infinite computing can be achieved through the confluence of quantum mechanics and computational designs.
In this session, Carl Bass, had invoked this term to denote the scale at which the computational powers of machines are growing. He compared the rapid scale at which computers are growing in intelligence in comparison to us. It seems he had made a statement that the computational power of some of the most advanced machines of today is as powerful as the aggregate capacity for computation that existed a decade ago.
This comparison may panic at least someone among us after famous physicist Stephen Hawking has commended on AI. Apart from the hype around the race towards artificial intelligence and contradicting predictions, an interesting take away from such a discussion is our hope that even if machines become more intelligent than us, they should be kinder than us. More importantly we aspire that machines should be kinder to us. May be we can pause for a moment and think how kind we are towards the supposedly less intelligent species around us. Are we really showing loving kindness or compassion to the inferior species around us? At least let’s reflect about the kindness that we humans show towards people hailing from different races, cultures and religions. So connecting intelligence and compassion in the scope of AI may sound hollow and shallow.
Leaving the hypes and tropes of intelligence aside, let us come back to the theme called ‘infinite computing’. Yes, compared to our own past, today’s computers are way too smarter. Should we roll it back or destroy them or unplug them? No. It is sheer stupidity to stop the technological progress and our quest for imparting intelligence to machines. On the contrary, this is high time, we start pondering about the ethics we have left in the scientific and technocratic community.
What is driving our quest for artificial intelligence and deep learning machine? Is it the quest for enlightening ourselves and improving the lives as we proclaim. Or is sheer passion for more power and information to win a rat race in the market of momentary fluctuations, perpetual randomness and collective variances.
If it is a quest for knowledge and bettering humanity and nature, then learning machines will not be harmful to our pathways. We will find novel methods to harness the computational power and intelligence of these learning machines of our times. Like I mentioned in one of my earlier post in LinkedIn, we should learn from our learning machines with the right perspective.
Keeping these observations in perspective, I am thinking about the direction in which our machines are progressing. Infinite computing, hyper computation, Zeno’s machines etc are some models for future. We are not yet there. We are not yet sure if they are making sense at all. There are hypotheses floating around us. There are supporters and critics of hyper-computation. To those who are new to the construct called hyper computation, Wikipedia says “Hypercomputation or super-Turing computation refers to models of computation that go beyond, or are incomparable to, Turing computability. I suggest this construct should be revisited with the same vigor in reading and watching the ingenious life and works of Alan Turing, one of the foremost visionary in the field of computation and information theory.
We should move beyond the immediate benefits and perils of artificial intelligence or machine learning and try to understand the possibilities of constructing machines that can take computation to infinite scales and bounds. The simple reason is that, we are no way near the reality of intelligent machines. The hue and cry about the world ruled by machines is not grounded on rational investigations, rather based on irrational fear and ignorance about the history of knowledge, humanity and nature.
Photo Copyright: Flicker/PascalShodan