Home | Corporate World | The future: AI, Machine Learning & Analytics

The future: AI, Machine Learning & Analytics

  Most of us understand how dependent we are on technology. Companies particularly are painfully aware that emerging uses of technology like artificial intelligence and machine learning can make or break their future. As a future manager, it makes perfect sense for students to delve into this domain.  

Dr Suresh Srinivasan

Arecent report by global consultancy firm McKinsey states that today, artificial intelligence (AI) is packed into everything we see around us, from products and services that we buy to most of the customer problems that are being solved by companies. The report says that AI is ‘the’ transformational technology of our digital age, and its practical application throughout the economy is growing apace.
According to management consulting firm Boston Consulting Group, rapid advances in machine vision and language processing are becoming organisational cornerstones. Companies are striving to strike a balance between people’s skill sets and machines, thereby radically enhancing their competitive advantage. This results in better understanding for the company, lets them address customer pain points, and at the same time do all of it with more efficient processes.
Bain & Company, another of the Big Four consultancies, says that the next phase of automation, based on machine learning, artificial intelligence and advances in robotics, will affect more than 50% of today’s workforce. The firm further cautions that the pace of such technological adoption will most likely become more rapid in coming years.
So how significant are artificial intelligence, machine learning and analytics (AIMLA)? Are these only management jargon or catchphrases that help consultants make tonnes of money? Is AIMLA here to stay, gradually becoming inherently ingrained in the way companies are run, or is it just a passing phase like many other technology jargon? As a student, do you really need to bother with what these are, and is this the right time for you to acquire skill sets in these areas, in addition to your domain expertise?

Here to stay
The mega-trend that is clearly visible today strongly indicates that AIMLA has crossed the inflexion point. You don’t need to be a genius to figure out that AIMLA is the future, and it’s already here! Of course, as far as organisations are concerned, there are leaders and laggards in AI implementation; some companies—and industries—are far ahead of others in using AI to gain competitive advantage.
At the end of the day, AI and machine learning are about accuracy in prediction, and many companies are fairly advanced, using large scale neural networks and algorithms that are learned and trained through large volumes of training data. The laggards will definitely need to catch up in the AI game in the coming days just to survive, let alone having competitive advantage.

How it works
Artificial intelligence is applied in machine learning. By training machines with adequate historic data, the algorithms gain an ability to automatically learn and improve from experience, without the requirement for any specific explicit programming. Data accumulation, writing algorithms, training the algorithms and prediction forms the essential basics of machine learning. These are widely used in healthcare, aviation and many other sectors of the economy. You’ve certainly seen or used products like Google Assistant, Amazon Alexa and Watson Assistant pioneered by IBM? Well, all of these extensively use AIMLA.

Not quite ubiquitous
Companies like Google, Amazon, IBM and Microsoft are miles ahead of the rest of the companies in their preparedness for AIMLA soaked products and services. However, even in developed economies like the US and Europe, many large companies are still only scratching the surface when it comes to AIMLA. Why are some companies far ahead of others in AIMLA adoption?
Management foresight, access to skilled workforce and the ability to allocate funds to AIMLA projects are some of the reasons that help some companies leap ahead of others in AIMLA implementation. Many companies are fighting for their survival in competitive markets today, and are unable to convince shareholders to invest for the medium to long term. To be fair, there are risks today in AIMLA investments to a certain extent—standardisation of the technology has not yet evolved, and there are still questions as to whether AIMLA projects should be done internally within the company or better outsourced to the so called consultants.

Implementation challenges
Successful AIMLA implementation requires an intricate mix of data science knowledge, skill sets and programming tools along with business and domain expertise. Companies need to pay a price to either attract such unique skill sets within the company or outsource the AIMLA project to worthy third party consultants. Naturally, such initiatives come with substantial costs, and drain management time away from their day to day initiatives. Hence, they take a back seat.
Given such a scenario, there is a natural tendency of the senior management of companies to procrastinate and delay investments in AIMLA, and resource allocations top such perceived ‘futuristic’ requirements. Such companies will soon realise that they have themselves out of the race and are behind the curve, and will be soon be forced to fall in line.

AIMLA and decision making
Managers at all levels take decisions. These decisions are required to solve problems that today are complex, in uncertain environments and a rapidly changing business landscape. Managerial decisions are regularly made in such dynamic environments, and no manager takes any decision with 100% information.
All of these mean that the only confidence with which managers take decisions today in the midst of such uncertainties is based on the comfort factor that machine learning and predictive analytics provide these managers. Decisions backed by data insight will become, or in fact is actually becoming, the new normal!

Nothing new
The whole concept of AI is not new, neither is it a wave of tsunami that has just arisen. AI has been in practice since the mid-1950s and had been revived in the 1980s and 1990s, when many of the techniques including in the areas of data mining and medical diagnostics were commercialised. However, back then, AI could never progress beyond a certain point as it was bound by the lack of hardware to process large data sets.
Then came the 2000s, which saw the emergence of cheaper hardware that could harness huge datasets. Especially with sensors become extremely affordable in the last few years, the ‘cyber-mechanical’ integration in the form of the Internet of Things (IoT) reaching new levels has now resulted in an explosion of data availability and capture. All of these, clubbed with data science skills sets, the capacity to write algorithms and managerial insights to solve business problems has now pushed deep learning and predictive analytics to the forefront in the last few years.

What you need to do
If we look at this mega-trend from a student’s point of view, and take a medium to long term view of your careers and the type of skill sets which employers will look for in you, the pieces will easily and obviously start falling in place. Data science knowledge and skill sets, programming tools like Python and natural language processing, clubbed with a strategic thinking mindset with clear understanding of business and domain specifics are skills that you need to steadily acquire and equip yourselves with in order to be marketable.
Gaining a strong grounding in statistical techniques will help you hone your business, functional and strategic thinking skill sets, as leadership at all levels require some amount of hands-on computing capability. Basic techniques like descriptive statistics and inference, moving to regression analysis and then further moving to advanced techniques like clustering, dimensionality reduction and ensemble learning techniques, slowly but steadily, builds a solid base for aspiring managers.
Some business schools are thinking ahead of the curve and have realised that creating appropriate skill sets in the field of AIMLA has a potential to transform the managerial landscape in the country. Business schools like Great Lakes Institute of Management, Praxis Business School, etc, have pioneered nurturing business analytics over the last five years. Now with specialised majors in the area of AIMLA, such institutes are also addressing the much needed raw material, i.e., the potential managers with requisite skill sets to power the AIMLA movement and advance the same globally!