As 2018 draws to a close, I see three factors that have grown in prominence – the shift to nationalism, climate change and the emergence of artificial intelligence – and imperil our children’s freedom. All three need to be tackled on an urgent basis. While nationalism and climate change are socio-political issues already, tools exist to handle these. However, AI threatens to very soon become a political issue, one that policy makers must start grappling with despite being woefully underprepared.
My thoughts about how terribly under-prepared we are, politically and socially, to the challenge of AI was trigged by an incident a few weeks ago at Bentonville Airport, Arkansas. Tired after a long day, I just wanted to get to Dallas where I had two very busy days coming up. I was flying American Airlines. Just before my zone for boarding came up, there was an announcement that they would check the hand luggage since there was no space in the overhead bins.
The lady at the boarding gate took my luggage and asks me if I am going to Atlanta. I said “No! I am going to Dallas.”. “But your ticket is till Atlanta”, she persisted. “Yes, ma’am, I go to Atlanta eventually after two days in Dallas”.
She says, “I am sorry sir, but I have to book your luggage to Atlanta!”. I protested, and her colleague at the counter piped in. “No sir, we cannot deliver it in Dallas!”. “But sir, I am getting off in Dallas and only take the flight to Atlanta after two full days in Dallas”, I spluttered. Both the folks at the gate were adamant that my luggage would go to Atlanta, regardless. While I was arguing, the lady prints the baggage tag and it says Dallas! Then she lets me go!
Without an apology she and her colleague carried on, as if nothing happened. But for the system, they would have had my luggage sent off to Atlanta while I would have been in Dallas. What I saw was a computer actually doing the right thing while the two humans were being stupid. In fact, my entire experience with American Airlines (AA) was revealing – as to how bad an airline can be, and how its employees can behave in a rude way, and be heartless and stupid at the same time. Even the international leg which I was unfortunate to fly in American was a horrible experience – with a surly flight steward who clearly hated her job and hated all of us. My fellow passengers too commented on it.
The lady at the AA boarding gate was merely following through on what she saw, without applying her mind. The Milgram experiment was a clear demonstration of how most normal people will follow orders from an authority figure, regardless of the consequences of following that order. Now imagine a conditioned employee who works in a culture that rewards conformity, obedience and efficiency, and cares two hoots for decency and kindness and humaneness. Now imagine that employee now taking orders from a machine. Will you think of the employee as the check against the machine, or would you look at him as an amplifier of the evil that is possible?
Far too many tyrants have been let loose on mankind while assuming that checks and balances will make them better in the seat of power. In the end, these checks proved inadequate to control them. They took over the institutions, and used them to deepen and strengthen their control over their citizens.
What is more frightening was that such tyrants were aided and abetted by an intelligent and powerful coterie that installed them as supreme leaders. Today, despite lessons from history, we are once again electing leaders who promise the world to their people, as long as they are given a free hand to enforce their will.
Would it be a stretch to imagine, one day in the not too distant future, a similar deference to an artifical intelligent being – real or imagined – that is elevated to a position of power? Also, one can imagine that access to this intelligence will be controlled by a coterie of unelected people. What would such a world look like? Will it be a kinder world, a gentler world, a world in which we can look forward to our children living freer, healthier and happier lives? Or will it be a world where heartless algorithms trigger the very same behaviour that sent millions to their deaths in the camps of Auschwitz, the killing fields of Cambodia, the gulags of Siberia and the famine of Bengal?
We cannot program kindness, mercy and goodwill into machines. That can only come from us. We need to recognise this fast. It is not about limiting AI, but understanding the limits of AI. Let us not leave AI to deal with life and death situations, and to decide on questions of morality and ethics. For this and more, we need policies to guide us on how AI is used. Policies that are shaped by well-informed debate, following the highest traditions of democracy. We can not depend on the milk of human kindness flowing through some corporate honcho’s veins to decide on what AI should and should not do. The final say is best left to humans who, unlike the two folks at the AA boarding desk, are actively trained and encouraged and supported for being human.
Note : This article was originally published on LinkedIn
Ramesh Dorairaj is consultant, coach and an author. He has 27+ Years of Experience consulting for Fortune 500 companies worldwide. He has groomed 50+ leaders. Has participated in 2.5 Billion $ worth of successful deals. He is a Certified Executive Coach at Marshall Goldsmith Stakeholder Centered Coaching, Certified Sales Coach and a Certified Proposal Coach.