I thank the Baroda Management Association for inviting me to deliver the Keynote address on ‘Artificial Intelligence and Impact on Human Resource Management.’ Technology has always interested me. So I decided to accept the invitation although the subject ‘Artificial Intelligence’ is yet to be understood well by common men like me. And much less, its impact on Human Resource Management. I congratulate Baroda Management Association for their choice of the theme as this conference will increase understanding of Artificial Intelligence.

It is a coincidence that I was reading about machine learning when Amit Karandikar called up and insisted that I deliver this keynote address. My interest in machine learning and Artificial Intelligence was fired by an incident. I was attending a meeting in which people were invited to discuss issues of ethics and corporate governance. Mr Manish Choksi of Asian Paints who is a wizard in IT was sitting next to me. He said that computers will soon have capability to decide ethical issues. I looked at him in total disbelief. I realised that Artificial Intelligence or AI is challenging Homo sapiens at a different level. The invitation to address this august gathering on Artificial Intelligence and Impact on Human Resource Management was too tempting to decline, though I realise that this field of knowledge is too recent to make accurate predictions.

In order to appreciate the impact of Artificial Intelligence on Human Resource Management, I will explore the nature of artificial intelligence, its capabilities, the impact on organisations, and what it means to HR Manager. I will also raise some questions which this seminar can explore further.

What’s the problem?

Artificial intelligence caught attention of the world when Deep Blue computer beat the world champion Gary Kasparov. That was in 1997. It was not too distant in the past, just twenty-one years earlier. The second game is the most discussed game in that contest for our purpose. Kasparov set up a trap by offering a ‘poisoned pawn.’ A poisoned pawn means an unprotected pawn; when the opponent attacks it he effectively falls in a trap which weakens his position and he is likely to lose the game. His assessment was that most computers, thinking mechanically, fall for it, but Deep Blue did not. Kasparov alleged that there was human intervention at work with Deep Blue. Apparently that was not the case. Subsequent analysis showed that both Kasparov and Deep Blue missed opportunities of perpetual check.

The incident sparked discussion on whether machines can reach intelligence levels of human beings or can even surpass human intelligence.  Incidentally, experts feel that Deep Blue did not possess anything like artificial intelligence. There are more powerful machines available now.

Genesis

The genesis of artificial intelligence is credited to Dartmouth Summer Project. It was here in 1954 that ten scientists convened a six week workshop on artificial intelligence. In the early days researchers built systems to counter the allegation that no machine could do a certain job. Today we have a robot, Sophia whose responses are amazing. Many of Sophia’s videos are available on the YouTube. Sophia’s quick responses and humour show the power of Artificial Intelligence.

The range of progress shocks us: from Deep Blue, which they say did not possess anything like AI, to humanoid Sophia in which visual data processing and facial recognition is used.

AI makes computers do tasks better than humans. It can be done only if the computers keep learning. This is where machine learning comes in. The intelligence in AI is associative intelligence. Algorithms help gain this association. As W. Brian Arthur says associations made possible by clever statistical methods using huge mass of data. As a result, certain pattern of pixel is recognised as, say, a cat or dog. This is precisely what makes a car run without a driver.

AI’s effectiveness is honed by two factors: prediction and judgement. Prediction and judgement are complementary skills. While humans will make decisions where judgement is involved, like ethical decisions, or diagnostic decisions in the case of doctors, the data will allow machines to get increasingly intelligent.

Add to this the possibility of singularity. What singularity means is that the invention of artificial superintelligence will abruptly trigger runaway technological growth, resulting in unfathomable changes to human civilization.

I am mentioning this because the stunning technological advance and the possibilities of superintelligence affecting us in ways we can’t comprehend, forms the basis of the discussion today.

And as I wrote my speech, Citibank informed me that I have been enrolled for voice authentication service. It means I can authenticate my identity just by speaking! Wow!!

There is so much drama in this story!

At the outset we have to recognise that there is a lot of drama in this plot. It has captivated human mind for a very long time. A man creates or tries to create something which will imitate him, he can control it but that entity goes out of control. Pygmalion is based on this theme. The Oscar award winning ‘My fair lady’ is based on this theme, and so also its adaptations in Marathi ‘Ti Phulrani’ as well as Sujata Mehta’s ‘Santu Rangilee.’

Let us now examine some questions relevant to Human Resource Management which come to our mind.

AI and Business Environment: Wages and Employment

The first question which comes to our mind is ‘what will happen to wages and employment?’ The machines will substitute human labour, thanks to AI. As a result, the wages will fall. There will be or may be areas where customers value work done by humans. [See ‘Superintelligence’ by Nick Bostorm]. These will be areas where humans will command a premium. Art work for instance. However we cannot say this with certainty. If the machine-produced art work is distinctly superior, humans may not hold premium.

Since the demand for labour will fall, the wages will take a nose dive. Perhaps wages will fall below subsistence level. India is experiencing this even without onslaught of artificial intelligence.

Which countries will go for AI first?

At this stage we should consider which nations will implement AI Technologies. Accenture Institute for High performance published a report in 2016. It says that “Managers in emerging economies seem more open to AI and intelligent systems than those in developed economies.” This is interesting. It means emerging economies want to use this opportunity to bridge the technology gap and also to close in on competition. Will India go for it early? It looks like yes, it will. And I shudder to think what will happen to people with falling demand for labour and also falling wage levels. It could be a big threat.

Unemployment and Obsolescence

There is a fear that the overall employment level will substantially reduce, jobs will simply vanish. The story of man versus machine is a decades old plot. Perhaps centuries old. It happened at the beginning of industrialisation. We saw it as early as in 1957 movie Naya Daur. Jeevan, the villain plies his bus for transport and jeopardises the livelihood of men. Unions in our country resisted loss of jobs. We are at a stage when nobody will be able to stop this onslaught of artificial intelligence.

I would like to remind you that the leftist resisted computerisation tooth and nail. Later in the mid-eighties computers were used for registration of delegates at the convention of CPI-M. The Telegraph says in its article on Oct 19, 2005 “…The protestors [belonging to Bengal’s left parties] said, the computers were a capitalist evil that would rob people of their jobs. The same leftists today showcase their government’s success in promoting the information technology sector.”

Technological obsolescence has worried working men as well as Governments. People and the Press talk of computers taking away jobs.  In early seventies, the Govt of Maharashtra constituted a committee to study the issue. The committee, of which the convenor was Dr VG Mhetras, who later became Director of Bombay Labour Institute [as it was called then] decided that some manual jobs will be lost, but new jobs requiring different skills will be created.

Technology usually creates jobs though in a different sector. Things could be different with machines becoming smarter than men is the warning of some scientists.

New Set of Jobs

Will we have new sets of jobs while old jobs vanish with automation and artificial intelligence?

All I can say with certainty is that the experts are divided in their vision of the future. Accenture carried out an in-depth study. To quote it ‘[H James Wilson, Paul Daugherty and Nicola Morini-Bianzino] identified the emergence of entire categories of new, uniquely human jobs. These roles are not replacing old ones. They are novel, requiring skills and training that have no precedents.’ The study further predicts that three new categories of AI-driven business and technology jobs will be created. They label them trainers, explainers, and sustainers. Humans in these roles will complement the tasks performed by cognitive technology, ensuring that the work of machines is both effective and responsible — meaning it is fair, transparent, and auditable.’

‘Trainers’ will teach AI systems how to respond to people’s questions with sympathy and depth. ‘Explainers’ will bridge the gap between technologists and business leaders. Explainers will help provide clarity, which is becoming all the more important as AI systems’ opaqueness increases. ‘Sustainers’ will help ensure that AI systems are operating as designed and that unintended consequences are addressed with the appropriate urgency. [“The Jobs That Artificial Intelligence Will Create”: H James Wilson, Paul Daugherty and Nicola Morini-Bianzino]

HSBC Bank has started using Artificial Intelligence to combat money laundering. I hope Punjab National Bank follows suit. HSBC Bank will automate the compliance process. Trainers, explainers and sustainers will help perfect the system.

So much for the different types of jobs which will be created. But the issue on the minds of everybody is – Will AI lead to higher unemployment. McKinsey’s Quarterly published an insightful article ‘Where is technology taking the economy?’ In it the author says that we are going to experience what Keynes called the ‘Technological Unemployment.’ The future of employment scenario seems gloomy! There are many who subscribe to this view. The report predicts that by 2030, as many as 800 million jobs could be lost worldwide to automation.

The Guardian article says that three types of jobs will be safe: these are [a] jobs that involve “genuine creativity, such as being an artist, being a scientist, developing a new business strategy”, [b] occupations that involve building complex relationships with people: nurses, for example, and [c] jobs that are highly unpredictable – for example, a plumber.

To sum up, new type of jobs will be created like trainers, explainers and sustainers. Some old jobs requiring creativity or complex relationship will survive the onslaught. But we should expect loss of jobs as Artificial Intelligence becomes increasingly common.

Challenges for HR Managers

The challenges before HR Managers are of humongous proportions. If his organisation is transitioning to the new technology, he will have to aid and assist the CEO in managing this discontinuous change. During the transition he will have to deal with redundancies and obsolescence. This presents a very emotion charged situation. He will have to train employees for the new jobs. It would be an uphill task because a completely different set of skills will be demanded.

These will be small problems as compared to the hard mind-sets he will have to deal with. What will change is the way people will perform their job. They will have to work with machines rather than operate machines because the machines will be constantly learning. Organisations will place greater value on collaboration and team work. Adapting to new technologies will be the competence in high demand. In a nutshell, we will have to change our way of working. It is certainly one of the most difficult transition to make.

The biggest problem in my opinion will be the nagging feeling that machines have taken over lives of human beings. Imagine the picture of autonomous cars moving on streets, plying their way responding to the distance from other cars, and stopping and starting at signals. That things can work without us gives pleasure in short run, but I guess this will be perceived negatively by people at work.

The feeling that intelligence is now outside of us is discomforting – at least several are going to find it so. I remember a story I heard in the mid-seventies. Automatic watches had just hit the market, they did not require winding the key. I know of a man who bought an automatic watch which he proposed to present to his son in law. But the young son in law declined saying that he did not wish to have anything that runs without his command! That’s an interesting message to the father in law. Humour apart, the discomforting feeling that intelligence residing outside us is quite universal. That might pose problem for both the organisation. It will also face the big problem of making work meaningful.

So far we operated on ‘Polanyi’s paradox.’ Stated simply, it means that we know more than we can tell. The resultant effect is that we created machines which can’t match our intelligence. But this very barrier will be transgressed with machine learning and AI. To people, working with machines which think and do better than us can be unnerving experience. How HR function can help people overcome such a barrier is a million dollar question.

Typically we see younger employees are more comfortable handling new technology than older. And they are usually at the lower side of the pyramid. Intranet and e-learning is not fully exploited by organisations because of older men at the top. I see possibility of high resistance to change, as well as obsolescence of older employees.

Culture and Ethics in Organisations

I would like to discuss impact of AI on culture and ethics. These are intermixed. The word ‘ethics’ is derived from the Greek word ‘ethos.’ It means both an individual’s character and a community culture. People believe that business ethics encompasses adherence to regulatory, legal, professional standards, keeping promises and commitments, and abiding by general principles like fairness, truth, honesty and respect. These are non-enforceable. The Institute for Global Ethics defines ethics as ‘Obedience to the Unenforceable.’ In life person what distinguishes one person from the other and in business what distinguishes one organisation from other is the practice of these principles. ‘Obedience to the Unenforceable’ is incidentally the title of an essay by eminent jurist Nani Palkhivala. [Adopted from ‘Business Ethics’ – The Economist]

Quite obviously what is right or wrong depends on circumstances and even then it is sometimes very difficult to decide. You would have heard the case of a trolley approaching fast on a track which would kill five men if allowed to go un-diverted. And if diverted it would kill one innocent person. What would you do? Sometimes it is right versus right conflict. Research tells us that there are four types right versus right dilemmas: they are Justice versus Mercy, Truth versus Loyalty, Short Term versus Long term and Individual versus Community.

How we resolve conflicts and dilemmas leaves a mark on an organisation’s persona; and these are not easy decisions. The issue is can AI learn to decide on values? The current state of technology seems to suggest that the answer is yes, though this is likely to be achieved a few years later. The essential question is what values to load. In other words it will be sometime before AI takes over making decisions which will leave an indelible mark on the culture of the organisation. This issue is important because it is organisation’s culture that every HR Manager is concerned about.

The Open Letter

I must bring to the attention of this conference one important development: I am reading out a few lines on Wikipedia:

“In January 2015, Stephen Hawking, Elon Musk, and dozens of artificial intelligence experts signed an open letter on artificial intelligence calling for research on the societal impacts of AI. The letter affirmed that society can reap great potential benefits from artificial intelligence, but called for concrete research on how to prevent certain potential “pitfalls”: artificial intelligence has the potential to eradicate disease and poverty, but researchers must not create something which cannot be controlled. The four-paragraph letter, titled “Research Priorities for Robust and Beneficial Artificial Intelligence: An Open Letter”[7], lays out detailed research priorities in an accompanying twelve-page document.”

And finally….

I have presented the issues affecting organisations and Human Resource Management. This is very futuristic subject and one that concerns all. The technology development is astounding. For a common man, there are so many new concepts to learn and so many surprises. There is so much crystal gazing involved, and it is complicated because the crystal is misty. “The trouble with future is that it arrives before we are ready.” The way out is to discuss various issues and learn from various perspectives. For this reason I congratulate Baroda Management Association for organising this seminar. I hope deliberations in this conference bring diverse perspectives on AI and impact on HRM. Ladies and Gentlemen, I wish this conference a big success.

Vivek S Patwardhan

PS: This was the Keynote Address I delivered at Baroda Management Association’s Round Table Conference on the subject on March 10, 2018.

I have been benefited by reading various articles published by HBR, Accenture and McKinsey’s. Also ‘Superintelligence’ by Nick Bostrom.