How am I doing on AI (Artificial Intelligence)?

March 19, 2019

Shejal Ajmera, Founder CrispIdea

Artificial Intelligence is currently the most disruptive force on the plant transforming variety of industries. However instead of talking about the size and its potential and its impact on jobs, we want to get straight to the point in terms of specific use cases across industries. These are the use cases already being adopted and something that business and technology leaders across all industries can start wide-spread adoption for their enterprises.

Now before we get into the use case, let’s get into the broader definition of AI first. Fortunately AI definition is not as ambiguous as Digital but it still covers myriad of technologies.

CrispIdea broadly defines AI as an interplay of technologies covering Computer Vision, Natural Language Processing, Machine Learning, Predictive Analytics and Virtual Agents.

Therefore let’s first define what are the all the elements that one could look at under the broader umbrella of AI:

  • Computer Vision – Image and video recognition technology.
  • Natural Language Processing – Includes speech recognition, natural language understanding and translation technology
  • Machine Learning – Software that can learn either through “supervised learning” where the correct answer is input or “unsupervised learning” where patterns are observed over a period of time and refined through inference. Data is critical to machine learning. The mechanism to get to machine learning will be through neural networks or deep learning
  • Predictive Analytics – Ability to predict outcomes based on machine learning
  • Virtual Agents / Chatbots – AI-driven agents in variety of applications such as customer support, chatbots, robo-advisers, personal assistants

Having understood these five broad areas that fall under AI definition, let us look at use cases industry-by-industry

Luxury Goods

  1. Personalization – Use of machine learning combined with additional social data
  2. Computer vision and augmented reality – Visualization in changing rooms
  3. Demand forecasting and optimized inventory management – ML based predictive analytics for demand planning, price optimization and inventory management


  1. Sharper targeting (e.g. location based offers) and high personalization
  2. Virtual agents and robot concierge service for guest check-in
  3. ML based product packaging for tour operators


  1. Recommendation engines – customer recommendations across music, video and e-commerce
  2. Programmatic advertising – buying and selling advertising inventory through an advertising exchange
  3. Predictive campaign forecasting – AI driven models to predict success of marketing campaigns


  1. Predictive analysis and markdown optimization – Align just-in-time ordering from suppliers based on predictive insights (from shopper buying behavior data)
  2. Credit and fraud prevention – assessment of optimal level of credit per customer, AI helps in detecting patterns of systematic fraud especially from new customers
  3. Returns optimization  – better knowledge of customer preferences to reduce high percentage of returns and also optimization of pick times within warehouse.
  4. Optimal routing for delivery services
  5. Next best offer – targeted promotion and personalized communications
  6. Checkout free stores – significant usage of computer vision


  1. Autonomous cars – OEMs have already put level 4 autonomous cars on the road
  2. In-vehicle content monetization – high level of personalization to monetize passenger’s time spent in the car
  3. Production cost optimization – fully autonomous production lines

Consumer Staples

  1. Targeted marketing and product development – Synthesizing product reviews across mediums
  2. Collaborative planning forecasting and replenishment – Analyze in-stock rates and reducing “out of stock” period
  3. Inventory optimization for short life products


  1. Analysis of subsurface data for exploration
  2. Predictive maintenance
  3. De-manning offshore operations


  1. Digitalization of customer journey – Replacement of retail banking middle and back office staff
  2. Improving risk, fraud management and compliance function
  3. Rapid loan disbursement – Leveraging customer behaviour data from multiple public, semi-private and private sources for credit assessment and disbursement of loan
  4. Virtual agents and robo advisors
  5. NLP based AI to analyze large amount of unstructured data (e.g. 10K filings, analyst calls)
  6. Automated fund management
  7. Risk based insurance pricing – getting more data streams on real-time customer behavior to price risks (e.g. motor insurance, health insurance)


  1. Drug discovery
  2. Safety monitoring through analysis of risks from large-scale claims database or large epidemiological databases
  3. Remote healthcare – Monitoring of patient condition in their homes or in remote settings
  4. Computer vision and collaborative robots for radiology staff and in surgical setting
  5. Machine learning for specific disease treatment e.g. Oncology


  1. Smart grids and Smart buildings
  2. Predictive maintenance and repair
  3. Smart homes
  4. Collaborative robots
  5. Autonomous mining, construction vehicles and drill rigs
  6. New product development

For details on each of the use cases, market size and potential and how companies need to respond to AI as an opportunity or threat, please feel free to refer to the following AI reports from CrispIdea:

Shejal Ajmera is the founder and editor of the technology research firm CrispIdea. She writes actively about Artificial Intelligence and is amongst the top rated technology sector analyst!

Latest updated blogs

Share this article on:

Shopping cart