The Industrial Revolution was when humans first overcame the limitations of muscle power. Often referred to also as the First Machine Age, humans during this period were largely complements to the machines. The Second Machine Age, which we are into currently, is mainly about complementing our mental faculties many times, using digital technologies. It is not too clear though whether humans will complement machines during this era or will be replaced altogether. Examples of both can be seen.
The defining moments when machines got the better of humans, occurred during 2011. In one of these instances, IBM's Watson defeated two leading players of the game Jeopardy!. In the other instance, during a machine learning competition, contestants were asked to design an algorithm that could recognize street signs which were rather indistinct and dark. Humans managed to correctly identify these 98.5% of the time. The winning algorithm did better than them, managing to identify these 99.4% of the time.
Since then, the question whether smart algorithms and software can replace managers, has gained primacy. In several instances, particularly when it comes to problem solving or finding answers to complex queries, machines have been doing consistently better than humans. Knowing whether to assert your own expertise or simply to step aside and let the software do the job for you is becoming a critical skill for executives in organisations which have been steadily adopting and integrating disruptive technologies into their operations. Yet, senior managers are far from becoming obsolete. As AI makes rapid strides, senior managers will be called upon to fashion the innovative new organizational forms needed to crowdsource the array of talent which is coming online all around the globe. They will also be called upon to emphasize their creative skills, strategic thinking and leadership abilities.
One of the biggest pitfalls for managers is underestimating the impact that data, combined with AI and analytics, can have on both their organizations and on society. One of the reasons for this is that the capabilities of disruptive technologies like AI and Predictive Analytics have been growing at an exponential rate and human brains often are unable to conceive of this. Two examples would help to illustrate this.
Google recently announced that it had completed the task of mapping the exact location of every business, every household and every street number across France. In the normal course, you would think that it would require Google to send out a team of 100 people, every day of the week, with GPS, to do this over possibly a couple of years. In fact it took Google just about an hour to get this done. What they actually did was to take their street-view database, containing hundreds of millions of images, and have someone go through them manually and circle the street numbers in a few hundred of these images. Then these were fed to a machine-learning algorithm which was told to figure out what was unique to the circled portions in the few hundreds of images, find them in the hundreds of millions of other street-view images and read the numbers it could find in those. That is what took about an hour. As would be evident, the increase in productivity and scalability by so many orders of magnitude, totally changes the nature of the challenges that the organization faces.
Let us look at another instance of how this same company, Google that is, does HR. It has a unit called the Human Performance Analytics group which takes data about the performance of all employees, what interview questions they were asked, what is the location of their office, how does that fit in with the organization structure and several other associated parameters. Then it runs data analytics to figure out what interview questions work best and which career paths are the most successful. Earlier, HR possibly never quite figured out some of these results obtained and possibly used anecdotal data and intuition to figure out some of the other answers.
Several proponents of disruptive technologies talk about how these can work seamlessly with humanity and not harm them. Some of these areas are:
1. Faster Onboarding
Typically new hires in an organization spend several weeks meeting new colleagues, going through documents and files and understanding and navigating internal processes. A knowledge graph, full of information gathered before the new employee is hired, can help cut down drastically on the time spent by the new employee in going through and understanding these processes. The knowledge graph can help answer questions like:
a) Who do I need to work with on the assigned task?
b) What were the meetings where this has already been discussed and will be discussed soon?
c) When is the next status meeting coming up and what information and updates do I need to be ready with for this meeting?
While these may sound simple enough, a knowledge graph will save the new hire plenty of time and help him or her navigate all the initial hurdles much quicker and easier.
2. More Efficient Workflows
Executives and managers can forget emailing meeting recaps and minutes. A.I. will help process decisions and assignments made in the conference room and add them to the knowledge graph appropriately. The technology for pattern matching and classification already exists, as would be evident from the 'Search box autocomplete'. In addition to linking decisions with assignments and steps needed to complete a project, A.I. has the potential to measure less tangible workplace influences and comb all communication to assess internal sentiments. Knowing what issues are being discussed most frequently, what concerns are being analyzed and how financial and emotional capital are being used would provide valuable layers of insights to managers. The Director of MIT's Media Lab terms the process of using intelligence as a network phenomenon and using A.I. to enhance, rather than replace human intelligence, as 'extended intelligence'.
3. Objective Performance Reviews
A recent Deloitte study found that only 8% of the organizations said that annual performance reviews were worth the effort required. While some jobs can be measured fairly successfully, what counts as success for most jobs is more subjective. Workplace politics, differing opinions and unconscious bias often take their toll on performance reviews. A knowledge web could capture every tiny detail of who proposed an idea in a past meeting and who managed the tasks to make it happen. While the need for people skills at work and human judgement in performance reviews will still remain, A.I, can help managers identify patterns in workers' strengths and weaknesses. This can, in turn, help managers make better assignment decisions.