Uncategorized

Collaborating with Robots. Yes, seriously.

“If you thought collaborating with the humans was hard, wait till you get assigned to a job that requires you to collaborate with a robot”

If we are to believe an increasing number of recent headlines then we are going to see robots coming: to take our jobs. That is old news to many. But for many others it comes as a threat that we are not sure how to handle. I’d put my money on collaboration being a big part of the solution. Collaboration just got a whole lot more interesting. It is not only a human thing. It will soon become a human and robot thing. Seriously, if you thought collaborating with humans was hard, wait till you get assigned to a job that requires you to collaborate with a robot.

cobot-750x400

We hear about technology that is encroaching on what has been exclusively human work. But increasingly the stories we read about have to do with robots taking over our human jobs. One University of Oxford study recently identified that an estimated 47% of employment in the USA alone will be impacted. Furthermore, along with the total loss of jobs, there is also potential for an impact on the wages of jobs that humans retain. As the authors of the paper point out, the issue of technology as a factor that makes some jobs unavailable to humans is several centuries old. Centuries before the famed Luddites even. The tension between our job security and advancing technology is well captured where the authors of the study state “The balance between job conservation and technological progress therefore, to a large extent, reflects the balance of power in society, and how gains from technological progress are being distributed.” Moreover, another report by Forrester indicates that automation will transform the workplace by displacing 22.7 million jobs in the US by 2025. The report predicted net job loss of 9.1 million, or 7% of the jobs in the US economy.

To me it is very interesting to imagine how humans will adapt to a world where non-human labour will impact. We already know that robots can write entire books; and in some cases better than many humans. There are many examples of what looks like robot work, which we can see slowly becoming part of our daily life. Self-driving cars, internet searches and medical diagnostics, for instance, already use the same technology. These are sometimes referred to as ‘weak AI’ but their scope will become much more prevalent in coming years with the influx of robots. The current application of robots in fields like medicine is a strong indicator of the benefits we can expect. As Dr Garnette Sutherland, Professor of Neurosurgery at the University of Calgary’s Department of Clinical Neurosciences and Hotchkiss Brain Institute says, “In 100 years humans won’t be operating on humans.” A veteran in robotic surgery, Dr Sutherland makes a point of explaining that the new field is also a product of collaboration between two disciplines; engineering and medicine. Another good example is how AI can be deployed to assist in analysing data and help doctors reduce costly misdiagnoses; for instance, analysing X-rays as in a recent international collaborative effort between an Israeli medical imaging company and an American health care provider.

The competition for jobs will not only be in areas of simple labour as many still seem to believe. The real value that robots will deliver will be in the knowledge economy. This point is well underscored by Microsoft CEO, Satya Nadella, who recently expressed a possible direction by arguing that “It’s not going to be about man versus machine, it’s going to be about man with machines.”

Non-human labour is not new. In fact, ever since the first civilisations emerged we have relied on it. Dogs were used to help us hunt for food, horses to travel, oxen to carry heavy loads, pigeons to convey messages, etc. Robots are no different. What might be different is our capacity to evolve. This is the area where we need to be much more creative and forward thinking. Robots, just like animals, have been employed to help humans. What we need to grasp is how they can best help us in areas where we need it most. And indeed there is no shortage of those. If robots can deliver better outcomes than humans then they should be used more. If humans want to keep the jobs they already have they will have to be better at them. I can think of many areas where humans complete a job that frankly leaves a lot to desire.

The centuries old attitude that looks to prevent technological advancement when it threatens to disrupt jobs is hardly sustainable. While we may still rely on it as part of a broader approach to keep things in balance, we need to also include two other parts to the mix. One is increased competitiveness by human labour, and the other is collaboration with robots. Already I can hardly wait to be served by a robot in restaurants, department stores and many other areas. I feel at times a robot’s smile would win me over more than chronically poor customer service. However, if customer service improves I might be less inclined to want to speak to a robot.

A major issue is how to balance collaboration between human and non-human labour. Collaboration is a good strategy for keeping costs down. Similar aims can be achieved if we look to upskill people to be better at co-working. A joint decision-making and even meaningful discussion between a robot and a human can in reality prove to be a very satisfactory outcome. Creating robots that are capable of creativity and imagination on par with humans may not be an unattainable task as such. However, the cost of doing so may be a limiting factor. Therefore pairing people and robots could lead to levels of innovation and job creation we never thought possible. All this leads to one point: a knowledge economy is the way to go.

So if collaboration is an option, how prepared or willing are we to engage? That very question was posed in a recent article by Dr Danushka Bollegala from the University of Liverpool. Dr Bollegala expands his point by asking how prepared are we, not just to work with but also to work for, a system that is powered by artificial intelligence. The answer may be complicated but, more importantly, much more relevant to our near future than we might be prepared for.

I find this a very interesting question for a number of reasons, collaboration being one of them. I wonder if the people who already embrace a collaborative way of working will have less trouble in expanding their collaboration with robots than those who resist collaboration for whatever reason. I often repeat Roadmender’s tagline ‘the future belongs to collaboration’ and I know many share that view. But perhaps it may still be hard to accept for those who never connected the dots. Collaboration is not exclusive to the human workforce!

Dr Bollegala raises another point that I find of critical importance if we are to make collaboration between robots and humans work; the role of emotional intelligence. As I asserted earlier, collaboration can be one factor that makes humans more necessary in value creation, thus eliminating some of the fear associated with intensification of robot use. Robots make mistakes and judgements can be enhanced by human capacity to add emotional intelligence; going beyond the pure algorithms that automated systems rely on. This can be further enhanced by an additional factor that gives humans a competitive edge; agility. Some notable manufacturers, like Mercedes Benz, have already come to this precise realisation and have in fact reversed the full automation process by replacing some robots with humans.

Returning to the question of collaboration between humans and robots is vital as we now know that the age of such arrangement has already began. Enter co-bots; or collaborative robots as they are properly known. Co-bots are in fact on the rise and have been increasingly deployed to work with humans. Early signs suggest that in some industries, as co-bots make work easier, humans are adapting well. Perhaps more encouraging is the fact that collaboration has proved to be better for these businesses as the co-bot/human collaboration has proved to be more productive than either system on its own – one study had the results at an amazing 85% more productive!

The reality of a new arrangement is not lost on the labour unions either. While many have yet to start looking into the issue in any real depth, some have acknowledged it. The rapid rise of robots, co-bots, AI and technology in general has not yet been fully understood. Research needs more data and for that we need time. There is a significant role to be played by business leaders being engaging, not just as managers but also as thinkers and strategists, because a lot of solutions will depend on judgements that may not be possible unless we accept risk and the simple fact that a lack of data is just part of the puzzle. It is in that context that I see collaboration as a critical skill that needs to be taken into account. From what we are seeing at present, the likely scenario will not be an extreme one where humans are working for machines or vice versa. As researchers at Forrester indicated in the previously cited report, the future is in collaboration between robots and humans. That future is already being tested across the globe. Take for instance work in Germany by the Technology University in Dresden where researchers have developed wearable technology for humans to train robots; designing intelligent clothes that enable humans to both control and, much more importantly, teach robots used in lightweight design and construction. These are cheaper robots that can be used by small businesses. They key advantage is that they can be taught to ‘do complex workflows and also repeat them’. The idea of robots learning via natural language processing is now a reality and that opens up big possibilities for businesses to deploy robots along with humans. And the idea of robots being a part of a value creation chain is no longer a matter for large companies alone: the simple fact is that real benefits will flow across business of all types and sizes.

Worries about the future are now a standard topic of many conversations. The sense of constant change creates what I have termed ‘unresolved uncertainty’; a sense that uncertainty is more than unknown: it is also hard to place in any specific narrative that helps humans mitigate anxieties associated with the future. Rationally speaking, humans do know that life is full of uncertainties. But we also possess the ability to imagine a future, to contemplate a narrative that offers hope and a sense of purpose and direction for our actions. Unresolved uncertainty refers to a sense that those tools may not quite work as well as they used to do. And this is largely due to an overwhelming sense of disruption not permitting us time to adapt.

Taking those concerns into account and looking at the work that has been done to date, and continues to evolve rapidly in the way some industries and individual businesses are adapting, it is clear that there are many more scenarios that are not fully explored. When we look at the long term options that will allow businesses to compete, innovate and grow without raising centuries old concerns about mass unemployment and all the side effects it creates, it is clear that we are going to need to embrace collaboration more intensely. Collaboration between humans has been adopted at an ever increasing rate over the past two decades but it will be the robots that will make ‘necessity’ the point of no return. We will soon come to realise that robots are not only a focus for select industries and large businesses. Their impact and relevance will be widespread no matter the type and size of a business. Those who respond faster will gain the advantage. Preparing the workforce to collaborate is the first step. Pretty simple really.