This was orginally published on Medium, using this space to constantly update use cases
With the evolution of Artificial Intelligence(AI), there's been a lot of potential business use cases which not only automates and optimize businesses process but also benefits end-users in terms of better engagement experience, personalized recommendations and so on. Here we're going to look into each industry(the majority of industries) and see how AI is impacting businesses at scale. source
- AI: The Big Picture
- General use cases of AI: Business Agnostic AI
- AI-First Companies
- Deep dive into industry-specific AI
Artificial Intelligence and Data Science are multi-disciplinary and are a wide field that aims at replicating human intelligence or at least a specific part of it through machines.
As mentioned earlier, Data science and AI are too broad, and to achieve the expectations of AI i.e "giving machines the ability to learn and perform specific tasks on their own" we have Machine Learning which is a subset of AI.
Machine learning needs 3 major components to achieve its goal:
- Data: This is the most important part of any AI pipeline and it can be anything like Quantitative, Categorical, Text, Image, Audio
- Features: Features are important pieces of data that help us in solving the task at hand
- Algorithms: Algorithms are specific mathematical models that learn patterns from the features thereby developing the ability to generalize to the problem at hand.
Deep Learning is a subset of Machine learning which is inspired by the neural structure and functioning of the brain. Deep learning uses complex multi-layered neural networks to learn feature representations automatically and proven to be more successful for Unstructured data like Image, Text, Audio.
There are 3 major types of Machine learning:
- Supervised Learning(task-driven): Learns the relationship between input and the target variable
- Unsupervised Learning(data-driven): Learns, extract, and describe the relationship between data points
- Reinforcement learning: Agent learns to operate by exploring to maximize rewards using feedback.
Based on the nature and complexity of data and problems, various AI tasks and techniques can be applied. Here I'm just listing out some common types and each task itself a broad topic so if you're new to this give yourself some time to explore about various tasks that interests you,
- Computer Vision: Image Classification, Object Detection, Semantic Segmentation, Instance Segmentation, Face recognition, Image Reconstruction, and Style transfer
- Natural Language Processing: Machine Translation, Named Entity Recognition, Question-Answering, Conversational bots, Speech recognition, Text-to-Speech generation, Semantic search, and so on.
- Time Series Analysis: Understanding the past, Forecast Trends, Risk Assessment, Anamoly Detection
- Classification and Regression problems involving Categorical and Quantitative data which are usually tackled with Feature Engineering and task-specific algorithms
With a huge amount of data and the evolution of AI algorithms, AI is impacting each and every business. Here are some common use cases of AI that can be applied to typically any business,
- Conversational AI: This includes chatbots(typically text-based), voice assistants, and voice + text assistants. This is now everywhere from mobile phones (voice/virtual assistants) to customer service
- Recommendation engines: From Social media to e-commerce, and e-learning, recommendation engines are very popular and are personalized and targeted towards end-users
- Risk Assessment and Fraud Detection: Most businesses use AI models to predict future trends, detect risk and fraud, and fraud prevention
- Search Optimization: AI-powered semantic search engines brings about an enhanced understanding of user(searcher) intent, the ability to extract answers, and delivers more personalized results
- Mobile Applications: Every smartphone comes with in-built AI features like smart camera applications like image enhancement, gender and age identification, 3D facial recognition, smart intelligent assistants, typing suggestions like auto-correct, next-word suggestions. And most of the businesses are building AI-enabled applications to optimize the user experience.
There are companies that advance on the use of AI and are focused on data collection, predictive modeling, and getting quantitative feedbacks from users. And not to mention most of the AI world depends and operates based on open-source research, data, algorithms from these companies.
Companies that own most of the user's data and having better prediction models will dominate the industry and win the digital economy.
Here are some companies that have declared themselves as AI-first and actually are(most companies/startups use AI-first as a buzzword - beware of them 😐),
- Google(Alphabet): They not only have good company merger and acquisition strategy, but they also have the world's best data acquisition strategy. With their search engine, chrome, android, youtube, maps, email, photos, they've live feed of data from every individual(maybe not in China😝). And, Google is one of the companies to invest early in AI Research and made a lot of breakthroughs in AI Software and Hardware. Now they're selling great Enterprise AI products/solutions with their Google Cloud Platform
- Amazon: Amazon is not just an e-commerce company anymore, they're into Cloud Computing, Live streaming, AI products, Robotics, and so on. They also have great data and company acquisition strategies. They have great recommendation engines, ai assistant, cloud ai products/services, and robotics products
- Facebook: Most controversial company in terms of privacy, and manipulation. They not only own platforms like WhatsApp and Instagram, but they also own crucial data of users from their platforms. Facebook AI (FAIR) has made great contributions to the open-source AI community
- Microsoft: They own Linkedin, Skype, Github, Microsoft office suite which are great sources of data, and Microsoft being a great player in the Enterprise Cloud Computing Platform has done a lot of breakthroughs in AI and now providing Enterprise AI solutions at scale.
The above companies are not the only AI-first companies, I have used them because they're more popular and easy to convey their data acquisition and modeling strategies. There are other companies that are also AI-first and popular like Tesla, Uber, Netflix, Alibaba, etc..
Let's dive into each industry and see how AI facilitate them and impact their business process and customers. I have listed the majority of industries and they're in alphabetical order so that you can easily find the industry you're most interested in.
- Automotive
- B2B Sales and Marketing
- Consumer Marketing
- Digital Commerce
- E-learning, Learning, and Development
- Engineering and IT
- Finance & Operations
- Healthcare
- Human Resources and Recruiting
- Industrial and Manufacturing
- Legal and Compliance
- Logistics and Supply chain
- Online Security and Risk
- Productivity
- Sports
- Telecom
The intention of this article is not to list every possible use cases in each industry instead, I'll focus on highlighting the top business use cases of AI and give some insights on how businesses are leveraging AI and leave it to you to explore more business problems that can benefit with AI
Self Driving Cars come into everyone's mind when they hear about AI in the automotive sector, but the industry is also working on AI applications that extend far beyond - Engineering, Production, Supply-chain, Customer experience, and so on.
- Autonomous Vehicles: There are 6 levels of automated driving ranging from 0(No automation) to 6(Fully automated), we've not yet reached level 3(Where the driving environment is monitored by AI) in production. But we do have some great features like Lane Departure Warning(LDW), Lane Keep Assist(LKA), Automated Emergency Braking(AEB), and Highway Co-pilot. Companies like Tesla, Apollo, Waymo and not to mention tech giants like Uber, Google, Amazon, Samsung are leading in this research
- Vehicle Design: The future of Car designs lies with Generative Design and Companies like Nvidia are using advanced Generative models, real-time ray-tracing to accelerate Vehicle design Workflows
- Production Line: Use of Robotics and AI in the production line is not specific to the automotive industry but companies like Tesla are using a fully automated assembly line to boost their production
- Quality Control: AI helps in Detecting minute damages and cracks in products which is a tedious, time-consuming, and error-prone task for humans
- Car Dealership Experience: Treating users with a personalized digital experience and digital engagement gives an enhanced customer experience
- Automotive Insurance: Damage analysis and automated claim assessment and processing speedups insurance processing
- Passenger Experience: In-car infotainment systems are now powered with AI and delivering passengers a pleasant experience. Example: AI-powered personalized audio search.
Sales and Marketing are essential elements for every business and there's been a lot of Analytics and AI-driven products that improve them.
- Account Prioritizing: AI-driven Analytics on the buying patterns of various customers gives great insights and helps in framing 80/20 rules to prioritize specific accounts
- Story Telling: Using AI and advanced analytics to draw insights and create stories from CRM data like Salesforce data
- Content Delivery: User targeted content-delivery systems are very helpful in generating leads
- Human-Friendly CRM
- Sales forecasting and Sales pipeline management
- Common use cases like AI-based meeting schedulers, negotiations.
AI is a great tool that can understand individual users and can personalize everything to the end-user at a very low cost thereby improving customer experience and customer relationships.
- Optimized search: Personalized search experience is something that every user would enjoy at the same time it helps businesses make more money
- Sales Analytics: Analyze Sales call and chats to improve selling experience and generate more leads. This is also useful for lead follow-ups
- Customer Interaction and Interaction Analytics: Digital engagements are a powerful tool for a brand to sell emotions and real-time analytics during a sales call gives immediate feedback and improves selling
- Customer Support: Every business now use ai-driven bots to attend and answer customer queries, this reduces wait-time and also reduces business operation costs
- Transcription and Translation: Real-time transcription and translation facilitate businesses to attend customers from various geo-locations from a centralized office.
We've already seen about b2b marketing, while the use cases are almost similar there's a big difference in the targeted audience which introduces its own complexity and flexibility.
- Marketing Funnel Solutions: Analyzing brand health and marketing campaign analytics are now powered by AI and speeds up this process
- Aid Words/Content Generation: AI is more powerful in text-analytics, using AI to aid content generation or to generate specific-words can help businesses win customer emotions
- Targeted Advertising, Personalized recommendations, Conversational Analytics are areas where AI is widely used.
Digital commerce has changed the way we buy and sell and they're the earliest sector to use AI-powered analytics to improve their businesses at scale. Recommendation engine, personalized user experience, and Virtual assistants are such use cases, but there are several other potential AI applications empowering digital commerce,
- Search: Image-based similarity search and NLP based semantic search are crucial parts of digital commerce
- Fraud Detection: Analyzes the potential risk of each transaction to detect chances of frauds, and anomalies like bulk transactions from stolen cards, multiple transactions in a short amount of time, transactions from unusual locations
- Sales and CRM Analytics: Identifying customer segments, targeted advertising, Customer retention
- Integrating products with everyday-use household items helps in better analytics
E-learning has great advantages over traditional learning and with the use of AI, both e-learners and platforms are getting better, smarter.
- Transcription and Translation: This is not only making content accessible to all geographies, but it's also giving accessibility to the disabled 😍
- Content Analytics and Personalized learning experience
- Automated Questions generation and Answers Evaluation
- Test monitoring: Using cloud and edge AI technologies to monitor candidates and detecting frauds. Example: Face verification, Eye movement tracking.
Still, most of the businesses use AI as a tool to facilitate their business process, Software engineering is no different.
- CI/CD Optimization and Log Analytics: AI-driven analytics help companies gain more insights about usages in production, improve software systems to scale, and reduce fail-overs
- Agile analytics and process-automation: Agile methodology requires a lot of chunking(division of tasks), and analytics for prioritizing and faster delivery. Using AI to automates the process and help teams focus on deliverables. Example: Good amount of time is spent in Jira/other tools for process-management, using AI to suggest and auto-label tasks, to add fix versions, link confluence, and feature branches to stories
- Quality Assurance: Automated functionality testing and integration testing is widely used and generates a lot of reports, using AI-driven analytics help businesses reduce the testing burden and time to production
- AI-powered IDEs: Use of AI for code auto-completion, code-optimization, suggestions on coding standards(naming conventions, better use of OOPs concepts) makes code development faster and easier
- Software teams develop AI-based internal tools that facilitate software development, testing. I myself have developed and used a few. If you're a software developer look around you, there will be plenty of boring stuff that can be automated and increases the productivity of your entire team.
Banking, Finance, and Insurance industries are early adopters of digitization and therefore predictive analytics. They also manage a huge volume of customer requests and by using AI/RPA they're able to deliver services at scale.
- KYC Automation: Using AI to verify customer identity reducing their processing time. Examples: Face Identification, Information extraction from documents
- Document Processing: Using Computer Vision and NLP to process documents reduces customer wait-time and operational costs. Examples: Claims and Invoice processing, Cheque processing, Customer onboarding
- Storytelling: Converting data into stories help users understand better, examples like spending analytics dashboard
- Algorithmic Trading: Algorithmic trading is so popular these days and has the ability to move the market. Currently, most of them are based on quantitative analysis but there are use cases where AI systems are used to understand the market news and sentiment and to execute trades at best possible prices
- Risk Analytics and Fraud Detection: Credit/Lending Analytics and Claims assessment are major processes
- Generation and Automation of Spending policies for Enterprises.
Similar to banking and finance, the healthcare industry is an early adopter of digitization and predictive analytics. AI is now impacting the healthcare business and had breakthroughs from diagnosis, prognosis, treatment, workflow management to medical research and drug discovery. There are plenty of use cases where AI-driven Analytics saved lives.
- Diagnosis & Prognosis: Use of Computer vision to aid predictions via Radiology and real-time analytics from IoT devices leads to a more accurate diagnosis, and also to forecast the effect of treatment based on the past medical history of individual and global data
- Patient Engagement and Tele-Health: Use of IoT and AI-driven analytics in patient monitoring and treatment is saving lives and making consultation and treatment more affordable
- Hospital Workflow Management: While most of the healthcare transactions happen via fax, the use of cloud fax makes workflow easier and secure. AI-driven Capture technologies (Information Extraction from fax documents, Documents routing, and search solutions) are reducing patient wait-time and time required for authorization, and claims processing
- Drug Discovery: Use of AI-driven analytics and software tools boosts Medicine research. Examples like identifying potential compounds. AI is also helping in clinical trails reducing time-to-market.
This industry involves a lot of record-keeping, conversational analytics, and trend forecasting to meet the demand.
- End-to-End Recruitment: Most Recruitment tools use AI at every phase like sourcing and shortlisting candidates, scheduling meetings, interview assessment, negotiation, and so on
- Performance Management
- Learning and Development
- Employee Retention Analytics
- Video Interview and Analytics
- Content Writing: AI helps professionals improve the writing of e-mails and other content because emotional feedbacks matter here.
Automation with sensors and actuators has transformed manufacturing industries for a long time and now with the help of advanced analytics and AI-driven automation, the manufacturing process is getting simpler and solves complex business problems
- Digital Twin: Virtual Representation of factories and products help in managing the performance, effectiveness, and quality of manufacturing assets
- Smart manufacturing with IoT
- Predictive maintenance: Device/Asset Diagnostics help business reduce unexpected down-time
- Quality Assurance: Assessing the quality of product and detecting minute damages made easy with AI-driven analytics
- Energy management and Waste reduction.
Legal industry involves more documentation and workflows which makes it easier to adopt AI to derive insights and process automation.
- Contract review: Using NLP to review contracts and generate insights reduces manual effort and saves more business hours
- Workflow and Legal Document creation: Design of legal workflows and documentation can be semi-automated with the help of AI
- Fraud detect and risk assessment
- Search: Faster and better-optimized search of similar legal cases, scenarios, and solutions
- Identity verification
AI is helpful in end-to-end processes of Supply chain management and helps businesses to meet global demands and optimizes storage and delivery problems
- Inventory and Distribution center optimization: Use of robotics and AI to fully automate the workflow
- Global Shipment Optimization, Shipment risk prediction
- Demands Forecasting and Continous delivery Experience
- Returns Management
From the workforce, commerce to healthcare, everything is now online. While there're enormous benefits it comes with security risks of privacy/data breach. The use of AI to detect, prevent security risks is more essential now.
- Remote workforce cyber-security: Most crucial elements of the business are now operated by remote workforce, recently twitter accounts of Bill Gates, Elon musk was hacked by targeted attacks on twitter employees. The use of AI models does not completely avoid risks but minimizes the chances
- Cloud Security: Every business now relies on Cloud solutions either public or private to serve their customers, use of better AI models to fight against cyber-attacks not only protects their data, but it also protects customer trust and brand health
- Mobile security: Everything now become portable like smartphones, laptop, tablets, securing these devices and the networks they're connected to is one of the most important things to protect end-users and customer trust.
Productivity tools provide many organizational benefits. These tools help individuals to work and collaborate effectively.
- For Teams/Enterprises: Analyze and work towards Key Performance Indicators, organizational goals
- Meeting Scheduling: Communicating with multiple members to align the meeting schedule and block calendars seems annoying and unproductive. There are a bunch of startups of working on using AI to automate this process just by having a virtual assistant in an email cc and it takes care of the rest. Tools like this are helpful for Sales teams, Recruiters, Internal teams, Consultants, and so on
- Presentation: AI reduces the tedious process of executives by facilitating them on tasks ranging from creating interactive dashboards to generating slides for a presentation
- Improve meetings: Transcription and Meeting notes generation, text, and in-audio search, file-sharing are some examples of AI use cases that improve meetings.
Given the advancement in technology, the use of AI in sports has more importance because of the availability of massive data, and analytics solutions that have the potential to transform the sports industry.
- Teams & Players: Analyzing complex metrics of potential teammates or players, workout & training patterns, and performance analysis and improvement are some of the areas where AI can make a huge impact
- In-Game: Making refereeing more accurate by using AI to aid referees decisions, improved match projections with AI-driven analytics and predictions
- Entertainment: Creating automatic highlights of games, better commentary & storytelling, optimizing advertising opportunities, personalized playlists in OTT platforms.
Telecom industry requires huge efforts on network monitoring, traffic management, and cybersecurity, these areas can be benefited from AI-driven analytics. And, there are some generic use cases like customer services, workflow automation.
- Network optimization: Optimizing network quality by continuously monitoring traffic based on region, time zone enables them to predict and detect anomalies and fix issues proactively using network scaling, Intelligent bandwidth on-demand
- Predictive Maintenance: Monitoring assets to anticipate hardware failure(Fault Prediction) and fixing issues can prevent faults
Hope I have covered the majority of business use cases of AI and will continue to update/revise these use cases. There are some new business areas that can be considered as AI-as-a-service like
- Business intelligence
- Data Science, and AI tools as a service: Analytics and AutoML tools
- AI service providers: Service-based companies that are mainly focused on Analytics and AI business use cases.
Happy Learning…!