The rise of AI-ML use cases post-COVID

With this piece, we take a look at how COVID-19 is changing the way industries will be operating in the future. We take a look at the AI-ML use cases; the improvement they offer over the current technology, specifically in Healthcare and banking. And, finally – what holds for these technologies in the future.

Let’s jump right in the piece!

COVID-19 is having profound business implications across various industries around the globe. Right from the shift in the demand curve to turbulence in the supply chain, COVID has shown the humans that their not-so-perfect world was just a pandemic away to be in shambles. While industries like Automobile, Travel & Hospitality, Oil & Gas are still recovering from the onslaught caused by the disease, various other industries have started eyeing the future. The tomorrow that lies in the hands of technology and more so with AI and ML-driven mechanisms.

AI and ML are the buzzwords right now and are often used interchangeably even though they are very different. In layman terms, Artificial Intelligence(AI) means creating machines smart enough to carry out tasks in a similar way as a human does whereas Machine Learning is a subset of AI in which a machine learns from the data feed without prior programming.

The capability of AI & ML in minimizing operational costs evolving workflow efficiency, and automating services are attracting organizations to leverage the technology at dispense. Technologies like Robotic Process Automation (RPA) are already helping employees by completing their computer-based mundane tasks. With most of the employees working remotely, it helps them focus on more humane tasks. Developments in AI applications such as natural language processing, data analytics will not only bring the businesses back-on-track, but they would also give them deep insights about customer behaviour to deliver effective services.

Related: Take a look at Neuromarketing
How AI improves performance – Use of predictive and prescriptive analysis

In business analytics, predictive and prescriptive are the two most important methods to study. Research suggests that the prescriptive and predictive analytics market (valued at $8.14 billion in 2019) is expected to develop at a CAGR of 22.53% to reach $27.57 billion by 2025.


Predictive analysis uses AI & ML based algorithms to build predictive models. These models are then used to analyse data and figure out underlying patterns to foresee future consequences. Nike stores in New York during the pandemic leveraged the 2019 acquisition of demand sensing technology. The predictive modelling-based technology warranted two things – gauge the approximate customer demand during the pandemic and ensure the correct planning of inventory so that shoes are delivered to the customer within a day or two after they order. Not only did it increase Nike’s digital sales (a majority of which came from D2C channel) Q-o-Q by 83% but has also cut the cycle time of orders by 50% in some cases.

Apart from the regular-structured data such as volume and inventory, firms and retailers are now focusing on their customers and employees who are generating unstructured public data which is valuable in maintaining safety and best practices across their stores and warehouses.

The prescriptive analysis focuses on finding the best decision-to-take in a scenario based on the data available. Though it is a combination of past data (descriptive) and future outcomes (predictive), it emphasizes less on data monitoring and more on actionable insights. A recommendation engine is the best example of how prescriptive analytics can be used by machines to predict the choice of a customer among a list of given items.

Etihad Airways uses prescriptive analysis to take critical business decisions. Over the pandemic, they are devising their network plans based on datasets like brand equity, market shares, GDP, historical performance, the performance of competitors, etc. Then the demand forecasting is done keeping the categories, class, origin, and destination in mind and further broken down into pricing segments. Also, the channel of booking is considered. Using all the aforementioned factors, Etihad decides where to deploy their fleets and what routes to cover up.

Though a little more complicated than other forms of data analytics, prescriptive analysis is worth the effort as it not only impacts the bottom-line it also optimizes the customer experience and helps companies make smart decisions.

Renewed Focus on AI/ML

The effects of COVID-19 are not going to be different from that of the Great Depression of 1929, or the Dotcom bubble of 2001. With increasing online traffic and work from home becoming a norm, we can expect significant changes in consumer behaviour. The application of AI/ML will help the companies adopt and identify these trends faster.

For example, advanced robots with supervised learning algorithms will be able to manufacture essential items 24*7, and as per varying demands, to prevent the supply-demand imbalance which the world faced during initial stages of the pandemic.

  • Federated Learning, wherein different parties employ their algorithms to learn from the same data set without any compromise on privacy, witnessed an explosion in 2020.
  • More than 1000 research papers were published on AI in the first six months of 2020, compared to a mere 180 in all of 2018.
  • The demand for A.I. talent continues to outstrip the supply, with the rate of job postings accelerating 12 times faster than the job views.

To cater to the changing dynamics, a renewed focus with higher investments is expected in the AI domain by global companies. It set to increase at a staggering CAGR of 33% from the current $50.1 Billion to more than a $110 Billion by 2024. The top sectors along with the retail which will see an exponential increase in spending would be the Banking and Healthcare sectors.

The rise of AI-ML use cases post-COVID

From the current figure of $463 Million, the spend behind Artificial Intelligence is set to increase to more than $2 Billion by 2023. Artificial intelligence is playing a vital role in the current pandemic, with firms like Google DeepMind and Alibaba developing tools to diagnose the virus evolution, track the geographical footprint and also predict its protein structure in order to expedite the cure.


Big pharmaceutical companies like Pfizer and Sanofi are leveraging the current pandemic and artificial intelligence to predict future occurrences and thus develop a timely cure. Artificial intelligence has a high potential in terms of developing Virtual Assistants that can automate roughly 70% of the interactions between the doctor and patient, thus utilising doctors only in cases of an acute emergency and bringing in efficiency to treatment.


As of 2019, Artificial Intelligence was adopted by banks globally for a number of purposes, as shown below:

Use Cases of AI-ML banking

To successfully thrive post COVID-19, incumbent banks must become ‘AI-first’ organizations, with deeper integration of Artificial Intelligence in their day to day operations.

With this adoption, between 15% and 45% of customers of banks around the world are expected to cut down their physical visits to the bank, relying solely on online banking. This leads to vast amounts of data to be harnessed and used for targeted personalisation. For the banking sector, we observe the costs saved due to Artificial Intelligence adoption to be as follows, depending on the role.

Trends in Artificial Intelligence

Starting with the domestic scenario, our honourable Prime Minister addressed a virtual summit, RAISE 2020, focused on Artificial Intelligence. India is currently 10th among countries filing AI patents, the leaders of which are the USA, China and Japan. India lacks a robust AI research ecosystem, due to which it missed out on the investment by wealthy investors and billion-dollar giants like Google and IBM.

Thus, the initial onus is on the government to give a push, which it has done in the form of a project named AIRAWAT. It’ll tie up with NVIDIA and Intel to utilise their microprocessors to build supercomputers at institutes like the IITs, which will be at the forefront of AI research. They will also mentor AIIMS to develop AI in the healthcare sector. Thus, in the tech and healthcare domain, these institutes will act as incubators for start-ups. The inspiration for this comes from the Summit and ABCI facilities in USA and Japan, respectively, both of which are sponsored by the state.

Elsewhere, around the world, many frontline institutions and bodies have adopted AI to combat the spread of COVID-19. Currently, drones are being used in the USA to maintain social distancing, and these drones will be further tuned in to detect COVID-19 symptoms within a crowd, thus relaying real-time information and statistics to the healthcare bodies and local administrators.

In the latter half of 2020, there has been a sudden upsurge in medium and small enterprises around the world adopting the digital retail route. Such enterprises who were initially averse to online channels due to technological unfamiliarity, now foresee the two-fold advantage of personalization and safety during the pandemic. Cloud computing offerings with inbuilt analytics platforms are accelerating the conversion of these MSMEs into online channels for a higher competitive edge.

IoT and Industry 4.0 are now a thing of the past; the major upcoming trend is that of IoB, a.k.a Internet of Behaviours. This term was coined by Gartner wherein public and private sector capture and wrangle data to churn out experiences that augment or suppress consumer behaviour. A real-time example of this is health insurance, where the insurance company can monitor and view your daily exercises, calorie intake by fitness bands to adjust the premiums.

Related: Take a look at IoT and Industry 4.0 use cases

We are well aware of automation and its adoption by several companies for its daily tasks. AI can further increase the breadth by providing Hyper-automation. Traditional automation worked towards making repetitive tasks in processes automatic. Hyper-automation integrates AI with RPA (Robotic Process Automation), to dynamically discover business processes and create bots to automate the entire process. In effect, it will reduce up to 6% of the manager’s workload and eliminate organisational silos.

Consider a familiar office scenario: where there was an assistant to enter data manually, and HR to onboard new employees and IT to solve any network issues. With hyper-automation, however, all of this can be automated, and the staff can spend more time on actual value add activities to the firm. No more is a data entry, printing and scanning help needed. No more are the countless emails between HR, and IT needed to set up the employees’ first day. Thorough data-backed decisions will do all of this.

Looks like life will be a little easy in the near future? What do you think?


The co-authors of this piece are Shshank Pandey and Virag Shah. The piece is – without a doubt – rich with data. The authors clearly deserve a shout-out, do it by sharing on LinkedIn for good karma.

Join our community of 5000+ readers by subscribing to our newsletter.

One thought on “The rise of AI-ML use cases post-COVID

Comments are closed.