Applied AI & AI in business /

10 applicable areas for AI: Beyond chatbots and customer recommendations

April 6 2019/10 min read

Up until now, overall AI adoption has remained low (with about 20% of businesses using it), according to an HBR study. That’s all about to change, and deep learning is powering this change. The potential for deep learning to transform all types of organizations, industries and use cases is profound. We all know AI is growing rapidly, but exactly how is it poised to affect change in the real world?

In this blog post, we discuss ten applicable areas for AI through various proven use cases that could become viable and mainstream via operational AI.

Applicable area for AI #1: Healthcare

In the realm of healthcare, AI can enable more accurate diagnoses, more precise prescriptions, improved patient care and reduced spending overall. The ability to mine both structured and unstructured patient content (including medical image files, family histories and data on patients with similar histories) allows healthcare providers to diagnose patients more accurately and suggest tactical care management strategies personalized to each unique patient.

One area of healthcare that stands to benefit tremendously from AI is cancer diagnostics and therapy. For example, a critical part of treating brain tumor patients with radiotherapy involves “segmenting” the tumor, creating a segmentation mask to mark exactly where the tumor is located, in order to target it more precisely. The mapping is a critical part in this process, because it maximizes the efficacy of the radiotherapy while minimizing the risk. This is a very time-consuming task for a doctor to conduct manually, thus building AI models capable of perfectly accurate segmentations, could drastically improve cancer treatments where doctors’ accuracies are currently averaging 85% (+/- 8%).

Applicable area for AI #2: Retail and e-commerce

AI software uses deep learning to tag and organize large databases of images so they are easy to search and classify, an application that lends itself well to retail, where huge numbers of products require even huger volumes of images.

Image classification is an essential part of the customer offering for online retailers in particular. The ability to automatically tag an image of a product with the correct attribute labels enables a retailer to organize its product catalog and allows the customer to filter products down to specific attributes. Deep learning models can be trained to predict which type of clothing a specific image depicts — not only by color, size, shape and brand, but also by very specific features such as zippers, buttons and pockets.

McKinsey calls retail, marketing and sales “the area with the most significant potential value from AI,” and beyond image tagging, the applications are myriad in this field, not just in customer-facing use cases but in operations as well.

The applications of AI are myriad in the field of retailing

Applicable area for AI #3: Process automation/manufacturing

Another important manufacturing application of AI is process automation. Even without AI, automation is already pivotal to manufacturing operations because it makes operations more efficient and less reliant on human labor. But deep learning could power predictive maintenance by processing massive amounts of audio, video and other data to quickly identify anomalies and prevent breakdowns.

In the pulp and paper industry — one that contributes quite significantly to the GDPs of quite a few countries — AI can be applied to optimizing the process by which pulp is converted to paper, ensuring that the final product is higher quality by constantly adjusting process settings using automation. The data used in these AI efforts includes thousands of recorded high-resolution images of wood fibers in conjunction with other sensor information from the machines being used to process the pulp.

Few industries are better suited for artificial intelligence than the financial sector

Applicable area for AI #4: Financial services

Today’s financial ecosystem is already heavily enmeshed with AI initiatives. From “robo-advisors” dispensing investment advice to boosting fraud detection, where models actively learn and calibrate to security threats as they rapidly evolve in today’s world of constant security risk. Machine learning is particularly adept at flagging anomalies — unique activities or behaviors which may signal a security breach.

Applicable area for AI #5: Weather and power

Smarter use of grid power can help power companies run more efficiently, waste less power and reduce the need to add more plants. The forecasting of electricity demand is one major application, but there are others for AI.

Weather has historically been forecasted by modeling the physics of the atmosphere, whereas deep learning could use data and patterns to make predictions — rendering it far more accurate. By comparing predictions with outcomes over time, the model would continuously learn and improve. With the huge treasure trove of historical and current weather data at our disposal, weather forecasting is an ideal use case for AI.

Applicable area for AI #6: Logistics and supply chain management

We’re already seeing AI-powered robots and autonomous drones and forklifts run warehouses and order stock when inventory runs low. HSB predicts that AI could create $1.2-$2 trillion surplus in supply chain management and manufacturing. It could help optimize the routing of delivery traffic to improve fuel efficiency, reduce delivery times and make shipping, in general, more efficient.

AI can optimize routing of delivery traffic to improve fuel efficiency, reduce delivery times and make shipping more efficient

Applicable area for AI #7: Music & Art

Streaming music services have been using AI for a while now to provide personalized recommendations and curate playlists. But AI has even more evolved applications to creative fields like music and art, both in business and in artistry.

For example, deep learning can be applied to the process of curating music by training a content-based mood-tagging model that leverages recent advances in image processing with convolutional neural networks. Deep learning automates the process of assigning sentiment tags to tracks — a process that until now has been entirely manual and largely subjective. In the dance field, AI can be used to enhance the process of choreographing new pieces based on complex spatial movement data gathered from existing dances. The applications of AI in art are as vast as the human imagination.

Applicable area for AI #8: Agriculture

AI has many applications to agriculture, particularly in helping plant breeders pinpoint specific traits to help crops more efficiently use water and nutrients, adapt better to climate change or resist disease.

Plant diseases are responsible for 10 to 16% of the losses in the global harvest of crops every year — not to mention that, as the world population climbs rapidly, by 2050, authorities predict that we’ll need to increase food production by 50% to fulfill global requirements. AI has also been applied to the detection of specific crop pathogens such as wheat rust, using satellite image data to determine areas where the disease has infected crops in China, a nation that lost 29.3% of its wheat crop to wheat rust in 1950.

AI has many applications to agriculture

Applicable area for AI #9:  Self-driving cars

In many industries, AI techniques are being applied to shift paradigms within existing processes and products. But self-driving cars are one example of a product type that has emerged because of AI — something we could not have imagined a few decades ago.

On the surface, self-driving cars seem like a novelty, but the potential of autonomous vehicles is huge. It could be a valuable remedy for the current driving fatality statistics — in the U.S., for instance, 15 out of 100,000 drivers/passengers die in car crashes every day. Driverless car fleets could also solve another pressing urban problem — the space issues that parking lots, garages and traffic jams create when everyone is driving and parking their own car. Fewer vehicular deaths, lessened traffic jams and a decline in air pollution are three of the major benefits of applying AI to automobiles.

Self-driving cars are one example of a product type that has emerged because of AI

Applicable area for AI #10: ...Any place a business is already using analytics

“...modern deep learning AI techniques have the potential to provide a boost in additional value above and beyond traditional analytics techniques ranging from 30 percent to 128 percent, depending on industry.” — McKinsey & Company

These are all excellent and actively pursued use cases, but in actuality, many areas of business or beyond could benefit from AI. In your own organization, look for places that AI  could provide more value. The opportunities for AI to transform business are profound —  assuming that organizations can surmount the obstacles of AI using operational AI tools.

We believe that AI will prove to be the most important technology ever, and it’s impossible to overstate the impact it will have on health, food production, energy and resource management, business innovation and individual empowerment. AI will save millions of lives and transform billions.

Are you ready to get “AI ready”? Get started with the Peltarion Platform for free, here.


  1. Most of AI’s Business Uses Will Be in Two Areas — Harvard Business Review
  • Pauline Norden

    Pauline Norden

    Communications Manager

    Pauline Norden is the Communications Manager at Peltarion, with experience from working with social media, content and influencer marketing within the tech sector. She holds a degree from Stockholm School of Economics, and before joining Peltarion, she worked with heading up the online community at the private finance company Tink.

02/ More on Business & Applied AI