40 AI Use Cases: How Many Did You Know About?
40 and counting...
“The conditions [man] tries to adjust to are going to change, and change so darned fast that he never will actually adjust to a given set of conditions. He’ll have to adjust in a different way: he’ll adjust to an environment of change.”
— John W. Campbell
Open-ended, unspecific talk of how AI will transform our society, for better or for worse, ultimately obscures more than it illuminates.
Sandwiched between the breathless boosterism and despondent doomerism of the current AI discourse lies a cornucopia of practical use cases.
AI, in its current form, is a tool. At its core, a tool is a problem-solving device. Any grounded discussion of AI today must therefore be based on problems. What problems can AI solve, and what problems will it cause?
Below, we present 40 potential use cases for AI, across eight fields and industries, for your delectation.
We intend for this to be a living document and would love for you to engage with us on these ideas in the comments or on Twitter.
Let’s work together to make AI the best tool that we can.
Thanks for reading Infinite Loops! Subscribe for free to receive new posts and support our work.
1. Consumer Interaction
1.1 Chatbots & Virtual Assistants
AI can be used to develop chatbots and virtual assistants that, via natural language processing, can accurately understand and respond to human language. These chatbots may analyze user messages' sentiment, intent, and context, thus improving customer service and reducing the workload for human customer support agents.
What this could mean: Chatbots will be responsive to your messages rather than trying to force you down a default set of imperfect options.
No more Less swearing at the computer.
1.2 Intelligent home systems
AI-powered voice recognition technology can enable intelligent home systems to consistently understand and respond to voice commands from users, learn and adapt to novel concepts without human input, and be able to respond to complicated commands. This could make it more convenient and intuitive to control various devices and functions within the home.
What this could mean: An Alexa-style device that you can actually talk to naturally and which doesn't feel like it has been lobotomized.
1.3 Product Development
AI can assist in product development by analyzing consumer insights, market trends, and competitor data to identify new product opportunities, optimize product features, and accelerate innovation.
What this could mean: Better products, faster.
1.4 Virtual Try-On
AI-powered virtual try-on technology can enable consumers to virtually try on clothes, accessories, makeup, and other consumer goods, enhancing the online shopping experience and reducing product returns.
What this could mean: Less friction in the shopping process. Fewer returns. The chance to try on that questionable pair of half-price salmon-pink chinos.
2. Creativity & Coding
Here’s: "...GPT-4 is next-level. I have been using it to write programs in Python and Unity (programming languages I literally do not know at all!) by just telling it what I want in words: "I need to create an Amazon Echo skill that will flash my hue lights green and blue when I yell party. Can you create it?" It did, and now my lights flash blue and green. It told me what files to download, what websites to go to, and what to do. When there were errors, I just pasted them in and it corrected the code and told me how to fix problems. I didn’t need to know anything. You can code now. Try it.”
2.2 Digital Art
Generative AI can be used to generate digital art, either from scratch or by remixing existing images. This is allowing the unbundling of creativity from execution. You used to need to have both the ideas and the ability to execute, but now those who have great ideas can get the AI to turn them into a reality, expanding the artistic possibilities for billions of people while allowing experienced artists to have more stylistic fluidity since they are not constrained anymore to the medium or style that they trained on for years.
What this could mean: An explosion of new artists and styles. No more boring stock images!
AI can quickly parse and summarise large bodies of data, highlighting key points, identifying areas of uncertainty, and revealing repeated patterns.
What this could mean: Access to a personalized research assistant.
2.4 Video Creation & Editing
AI can be used to assist with video editing, suggesting cuts, transitions, and effects that can add a creative edge to the final product. It can also be used to generate AI avatars and environments.
What this could mean: A democratization of the creative process for filmmakers. An explosion of new stories and storytellers.
AI can be used to write first drafts, improve current drafts, test ideas, challenge premises, and test arguments.
It can also be used to generate lots and lots of ideas. The human then acts as a curator who picks the best and then develops them further (likely with further help from AI!)
What this could mean: A writing partner in your pocket.
3.1 Accessibility and inclusivity
AI can help create more inclusive educational environments by providing personalized learning experiences and support for students with disabilities, such as customized interventions, adaptive assessments, closed captioning, audio descriptions, text-to-speech, and speech-to-text conversion tools. It can also help identify students at risk of falling behind, allowing for early intervention and targeted support to help them succeed.
What this could mean: A more inclusive and equitable educational landscape
3.2 Adaptive learning
AI-powered adaptive learning systems can analyze students' learning patterns, strengths, and weaknesses and tailor educational content to meet their individual needs.
What this could mean: A more personalized and practical learning experience.
3.3 Grading automation
AI can help automate the grading of assignments, tests, and essays.
What this could mean: More time for educators to focus on essential tasks like lesson planning and student engagement.
3.4 Intelligent tutoring systems
AI-driven tutoring systems can provide students with real-time feedback, guidance, and support, simulating the experience of a one-on-one tutor and enhancing their learning process.
What this could mean: Affordable, accessible, personalized tutors for every child. A democratization of personalized education.
3.5 Virtual reality and simulation
AI-powered virtual reality and simulation tools can provide immersive, hands-on learning experiences that enhance the understanding and retention of complex concepts.
What this could mean: Imagine learning about Ancient Rome by stepping with your peers into a simulated version of the Senate House. Imagine practicing a speech in front of a virtual audience. All of this could be possible with AI-powered VR education.
4. Environmental Issues
4.1 Climate modeling and forecasting
AI can help improve the accuracy of climate models and forecasts by analyzing vast amounts of data from various sources, such as satellite imagery, weather patterns, and historical trends.
What this could mean: More accurate and actionable climate models and forecasts.
4.2 Environmental policy and compliance
AI can analyze vast amounts of environmental data, such as climate data, pollution levels, biodiversity records, and satellite imagery, to identify patterns, trends, and anomalies. This data-driven approach can provide insights and evidence-based recommendations to policymakers for more informed decision-making on environmental policy and compliance. AI can also assist in analyzing the effectiveness of environmental policies, tracking compliance with regulations, and identifying areas where additional efforts may be needed to achieve desired outcomes.
What this could mean: Less time spent wasted on unproductive bureaucracy.
4.3 Natural disaster prediction and response
AI can help predict natural disasters, such as floods, hurricanes, and wildfires, by analyzing weather patterns and other environmental factors. It can also aid in coordinating disaster response and recovery efforts.
What this could mean: Communities can better prepare for, and respond to, natural disasters.
4.4 Pollution control and monitoring
AI can help monitor and predict air and water pollution levels, enabling targeted interventions and policy changes to reduce emissions and improve environmental quality.
What this could mean: Better, more effective anti-pollution policymaking.
4.5 Smart agriculture
AI can optimize agricultural practices by providing insights into crop health, irrigation management, and pest control, reducing water use, fertilizers, and pesticides.
What this could mean: More efficient agricultural processes = more global food production.
4.6 Urban planning and smart cities
AI can inform sustainable urban planning by analyzing data on population growth, transportation patterns, and resource usage.
What this could mean: Possible ways to design more efficient and environmentally friendly cities.
5. Financial Services
5.1 Compliance and Regulatory Reporting
AI can automate compliance checks and regulatory reporting by analyzing financial data, monitoring transactions, and identifying potential violations.
What this could mean: Financial institutions can more easily ensure compliance with complex regulations and reduce compliance costs.
5.2 Document Automation & Processing
AI can automate the generation of legal documents, such as contracts, wills, and legal agreements, by analyzing templates, extracting relevant information from user inputs, and generating customized documents, saving time and reducing errors.
AI-powered natural language processing can analyze legal documents, contracts, and other legal texts to extract relevant information, identify risks, and support legal research and due diligence, improving efficiency and accuracy in legal processes.
What this could mean: Cheap, easy-to-access, customizable legal documents for all. Can you imagine a world without lawyers?
5.3 Fraud Detection
AI can analyze vast amounts of financial data, including transactional data, customer behavior, and external factors, to detect fraud patterns and identify suspicious activities in real time, helping to prevent financial fraud and protect against monetary losses.
What this could mean: Faster, more effective fraud detection.
5.4 Market Prediction and Trading
AI can analyze market data, historical trends, and other relevant factors to predict market movements, optimize trading strategies, and improve investment decisions.
What this could mean: Improved market predictions.
5.5 Risk Assessment
AI can analyze historical data, market trends, and other relevant factors to assess risk in investments, loans, insurance policies, and other financial products, enabling more accurate risk evaluation and informed decision-making.
What this could mean: Better risk assessment helps individuals and companies reduce losses and maximize gains when dealing with financial products.
6. Science & Medicine
6.1 Computational Protein Design
AI can predict the three-dimensional structure of proteins, a critical step in understanding their function and designing drugs, by analyzing their amino acid sequences and simulating their folding patterns.
In biology, function follows form. By being able to design proteins in silico, we should be able to more quickly discover useful molecules.
What this could mean: A revolutionary new understanding of human biology. The opportunity to discover new ways of fighting disease.
6.2 Drug Discovery
AI can accelerate drug discovery by analyzing vast amounts of data, including molecular structures, genetic data, and clinical trial results, to identify potential drug candidates and predict their efficacy and safety.
What this could mean: Quicker identification of effective medication.
6.3 Healthcare diagnosis
AI can analyze medical images and data, aiding diagnosis and treatment planning.
What this could mean: Imagine using your smartphone to take a picture of an unusual mole or other skin discoloration; have the AI analyze it against millions of samples of known (and potentially AI-generated suspected) skin cancers and report back to you in minutes.
6.4 Null Hypotheses
There is a benefit to knowing what doesn’t work. Unlike humans, AI doesn’t mind working on things that don’t end up working. AI could generate null hypotheses and machine-publish them directly in repositories so that other scientists and AI scientists could plug in and see what has already been tried to inform their own research.
What this could mean: More time for scientists to allocate to productive research.
6.5 Precision Medicine, Health & Treatment Plans
AI can analyze genetic data, patient records, and clinical trial data to identify patients likely to respond well to specific treatments, enabling more precise and effective medical interventions. It could also analyze healthcare data, fitness data from wearable trackers, and nutrition data to prepare a customized all-in-one personal health plan.
What this could mean: Personalized, bespoke medical interventions, treatment, and nutrition plans.
AI can reproduce studies 'in silico' to see if they replicate. Some bio-medical studies may be reproducible through computation biology, so AI can quickly and cheaply check if they reproduce without having to re-run a whole trial in the flesh.
AI can also be used to run variations on existing methods to find better matches/optimize outcomes in ways that a physical study doesn't have the resources to do.
What this could mean: More confidence in the scientific process. Less frequent publishing of ‘bad science.’
6.7 Scientific Visualization
AI can generate realistic visualizations and simulations of complex scientific phenomena, such as molecular interactions, quantum mechanics, or astrophysical simulations, to aid scientific understanding and communication.
What this could mean: Better scientific understanding of complex phenomena. Improved understanding of data sets and access to insights that simple statistical methods may hide.
6.8 Uncovering Fraud
AI can find anomalies in published research to uncover potential fraud or honest mistakes. For example, samples and results that have been tampered with may show patterns that would be very unlikely to happen by chance, or some results may be re-used, images edited by software that leaves artifacts, etc.
What this could mean: Less fraudulent research.
7. Social Care
7.1 Cognitive Support
AI-powered cognitive assistance tools can support individuals with cognitive impairments or dementia, helping with memory recall, task reminders, and decision-making support.
What this could mean: AI-powered brain training for the elderly.
7.2 Fall Detection and Prevention
AI-powered sensors and video analytics can detect falls or changes in movement patterns in social care settings, triggering timely alerts to caregivers for preventive measures and reducing fall-related injuries.
What this could mean: Quicker responses to falls and other medical emergencies. More peace of mind for non-live-in caregivers.
7.3 Social-Emotional Support
A sad reality for many elderly people is that they don't get as much care and companionship as they wish. There are simply not enough healthcare professionals to meet demand, and while for many families loved ones play that role, not everyone is in that position, and even the luckiest people don't get visited by family and friends every day. Because of demographic trends, this is likely going to become worse, not better.
AI-powered assistants and companions are potential tools to help with this situation in a way that is additive to real human contact (it should not replace it!). While not perfect, AI solutions have to be compared to the reality as it stands and not to an imagined world where everyone gets all the human contact that they need. If the alternative is a TV or a radio, having an AI that you can talk to, play games with (cards, board games, video games, etc), ask questions, or use it to assist you control other devices ("show me photos of my family on the TV screen" or "play my favorite music and don't forget to remind me about doing a video call with my grandkids at 7 PM today") would be a huge improvement. It should also be possible to have an AI trained on personal data so that it is aware of family relationships, contact information for loved ones, personal history, etc.
What this could mean: Immersive environments for elders that could significantly reduce loneliness and despair in a much more meaningful and enriching manner than leaving a TV set or radio on as their companion.
8. Transportation & Machinery
8.1 AI Powered Vehicles
Eventually, AI may be used to power self-driving cars.
Even if 100% autonomous vehicles do not become widely used in the short term, AI is already making vehicles safer with active and passive safety features, such as detecting and avoiding collisions, applying brakes quicker than any human could, helping detect danger in low-visibility situations, and designing safer vehicles through the AI-assisted design of parts and materials.
What this could mean: Safer roads, less congestion, reduced fuel usage, and more accessible transport.
8.2 Public Transportation & Supply Chain Optimization
AI can optimize public transportation systems by analyzing passenger demand, traffic patterns, weather patterns, and scheduling data to improve route planning, reduce wait times, enhance the overall passenger experience, and improve delivery efficiency.
What this could mean: More efficient transportation systems.
8.3 Predictive Maintenance
AI can analyze sensor data from vehicles and machinery to predict when maintenance is needed, helping to prevent breakdowns and optimize maintenance schedules, reducing downtime and costs.
What this could mean: Imagine having a car that tells you when it is going to break down!
8.4 Traffic Management
AI can analyze data from various sources, such as traffic sensors and GPS devices, to optimize traffic flow, reduce congestion, detect and predict traffic accidents, provide early warnings, and improve overall transportation efficiency and safety.
What this could mean: Less time spent wasted in traffic jams. Say goodbye to road rage during your morning commute…
We recommend subscribing toand , which provide excellent overviews of AI and its potential use-cases.
This is but a first draft. We have barely scratched the surface of potential use-cases. Which ones won’t work? Which ones will work? Which ones are undesirable? What are we missing, and why? Let us know in the comments or on Twitter and, over time, we will update this document accordingly.