Current Status of Global AI Adoption | CBA - Communication Business Avenue

Current Status of Global AI Adoption｜Case Studies in Six Industries

The reasons for considering AI implementation vary from company to company. For some, the biggest driver may be corporate branding: and the desire to be perceived as a competitive company that stays ahead of the curve. For other companies, the motive may be of a more practical nature – management may be under pressure to reduce costs, and so turn to AI as a potential solution.

For engineers, the opportunity to take advantage of machine learning technology may be a strongly influencing factor in the decision to implement AI.

However, when it comes to introducing AI, it may not always be clear how the technology can be used to bring real, measurable benefits to your company and business processes.

Communication Business Avenue has been helping some of the world’s largest companies to integrate call center systems and digital communication tools for the past 16 years. In this report, we’ll take a look at the current state of AI adoption in key global markets, before considering specific case studies of AI application. Find out how to make the most of AI tools that are indispensable for your company’s growth strategy.

What is AI?
Current Status of AI Adoption Abroad
Advantages of AI
Disadvantages of AI
AI case studies
7 Steps to avoid failures in AI implementation
Final thoughts

What is AI?

As it turns out, there is still no established definition for AI. However, the White Paper on Information and Communications published by the Ministry of Internal Affairs and Communications explains it as follows:

AI is understood as a broad concept that includes programs that operate in a manner similar to human thought processes, or information processing and technology that humans perceive as intelligent.

✅Simply put, AI is technology that uses software to reproduce some aspects of human thought.

AI is also called Artificial Intelligence, and sometimes the terms “machine learning” and “deep learning” are used together.

What is Machine Learning (ML)?

Machine learning is a type of computer science that allows computer programs to learn and improve all on their own. In the past, computers could only do what humans programmed them to do, but now with machine learning, they can behave like humans, gaining knowledge based on their past experiences.

Although the process itself can get very complicated, the basic concept of how machine learning works is not too difficult to grasp.

For example, let’s say we want to create a program that can tell the difference between carrots and potatoes. First, we give it a labeled picture of each. The program will then look for patterns in the vegetables and start to ‘remember’ them. It can then use these ‘memories’ to look at unlabelled pictures of carrots and potatoes and determine which is which all on its own.

Machine learning is often used in products and tools that we use on a daily basis. For example, social media sites use machine learning to create your feed based on your preferences and search engines use it to improve the accuracy of their search results.

What is Deep Learning (DL: Deep Learning)?

Deep learning is a subset of machine learning. Whereas machine learning algorithms leverage structured, labeled data to make predictions, deep machine learning doesn’t necessarily require a labeled data set. It can ingest unstructured data in its raw form like text and images and automatically determine the set of features which distinguish for example – a potato from a carrot. By observing patterns in the data, deep learning algorithms can discover hidden patterns of data groupings, without the need for human intervention.

For deep learning to work however, there needs to be a much higher volume of data with which to train the machine.

The Current State of AI Adoption　

How advanced is the adoption of AI in other countries? According to a survey conducted by Oracle Japan in 2019 on AI usage in 10 countries around the world, AI adoption is progressing in India, China, UAE, and the U.S.

⚠️India, with its abundance of IT talent, has the highest AI usage rate at 78%. Brazil also ranks high in Oracle’s 2020 study, with an AI usage rate of 54%.

Advantages of AI

What are some benefits that using AI could bring to your company? Here are six advantages you should know about:

1. Streamlining of business operations

Unlike humans, AI does not need to take sick leave or suffer from an occasional loss of motivation. It can continue to perform the tasks you set it, while maintaining a consistent level of speed and quality

2. Increased productivity

AI can be used to take care of monotonous tasks, leaving human employees to focus on the tasks that AI is not so good at (i.e training and mentoring employees, or providing emotionally intelligent customer service).

3. Cost cutting

Since AI can process purchase orders, receipts, etc., it can save both time and human resources.

4. Sales improvement

AI is good at analyzing past data and drawing conclusions that can be useful for future marketing efforts. Using AI in product development can create products that sell well, and using AI in ordering can reduce wasteful ordering.

5. Improvement of safety

AI can be used to detect problems with infrastructure and equipment as well as predicting illnesses.

6. Customer success

AI-powered avatar customer service enables 24/7 customer service. In call centers, AI can reduce waiting time by allocating customers without the need for AI to handle the customer service itself.

Disadvantages of AI

When introducing AI, it is important to know the disadvantages. There are four main disadvantages.

1. Introduction cost

When adopting a new technology, there is always an introduction cost. In addition, costs such as time and human resources to review the workflow to ensure efficient operation of AI are also required.

However, once the operation is on track, the investment costs can be recovered.

2. Reduction in employment

One of the concerns about the introduction of AI is that it will reduce employment. Employment of data entry staff and hotel room attendants, for example, who often perform simple tasks, may decrease.

However, considering the labor shortage, the introduction of AI is inevitable.

The introduction of AI will make it possible to utilize the personnel who used to do simple tasks in other operations.

3. Information leakage

AI, which processes a huge amount of data, is exposed to the risk of information leakage during its operation. However, the risk of information leakage is an issue that all companies must deal with, regardless of whether AI is operated or not.

4. Difficulty of risk management

AI operations have only just begun. It is not always clear what kind of trouble may occur during introduction and operation.

Consulting with an AI vendor with knowledge from overseas, where AI operations are already in progress, can help ensure appropriate risk management.

AI Case Studies

Let’s take a look at AI case studies in six different industries.

Financial Industry

According to a survey by the Economic Intelligence Unit, 77% of US bank managers believe that AI will separate the winners from the losers in the banking industry.

In fact, major investment bank Goldman Sachs (U.S.) is using AI to detect fraudulent activities such as insider trading. Japanese banking and financial services institution SMBC has developed the SMBC Chatbot in cooperation with Microsoft Japan.

Healthcare Industry

According to a study by Accenture, AI systems in healthcare are expected to “save” $150 billion annually in the US alone by 2026.

In the U.S., algorithms are being developed that can diagnose illness based on a simple recording of a coughing sound, and AI chatbots for healthcare providers are being introduced.

Apparel Industry

Israel-based startup Finesse uses AI to predict future demand. In New Zealand, interactive AI digital humans are used to serve customers, and in the U.S., AI is being introduced to learn customer preferences regarding customer service.

Japan based e-commerce company ZOZO Inc. has increased its price accuracy rate by 150% by using AI to price used clothing. Nano Universe Co., Ltd. has introduced an AI chatbot “OK SKY” on its e-commerce site to make it easier for customers to use its e-commerce site even late at night.

By making it easier for customers to use the e-commerce site even late at night, the company has been able to increase both the price-per-purchase and the number of purchases made per year.

Tourism Industry

Edwardian Hotels in the UK have developed an artificially intelligent, robot concierge named Edward. Edward is capable of assisting guests with hundreds of questions, including the nearest restaurants, bars, and railway stations and information regarding local sights.

Another example of the successful use of AI in the travel industry is the chatbot Rose, developed by the Cosmopolitan Hotel in Las Vegas. Rose has distinctive features including the ability to create an emotional connection with visitors, as well as delivering recommendations for restaurants and other local attractions.

Education

Efforts to use AI as a teacher’s assistant or virtual teacher have begun in the United States. Jill Watson, a virtual teaching assistant introduced by the Georgia Institute of Technology has been trained on an extensive database and is capable of responding to a range of queries. Jill can assist students with queries regarding the program, formatting academic papers and introductory emails.

Telecommunications Industry

In the U.S., 78% of call centers are using AI or planning to introduce it within the next three years. AT&T one of the largest telecommunications providers in the world, has begun leveraging AI to handle all of its online customer interactions.

7 Steps to Implement AI without Failure

There are seven steps to avoid failure in AI implementation.

Step 1: Identification of issues or intents.

Identify areas of your business where productivity is low or efficiency is low.

Step 2: Determine the issues in which AI should be implemented

Among the issues you have identified, divide the issues into “categories that can only be done by humans” and “categories that can be left to AI.

Perhaps at this stage, you may find that you can solve the issues not by introducing AI, but by simply using inexpensive Robotic Process Automation (RPA) tools that automate the work.

Step 3: Data collection

The more data AI has, the more accurate the analysis can be. Since it may be difficult to collect data on your own, consider outsourcing the task.

Step 4: Select AI tools

How to select an AI tool depends on whether or not you have in-house engineers. If you have engineers, you can implement AI using only existing platforms. If you do not have an engineer, then you should implement an AI tool for your company.

As explained in the disadvantages section, risk management is essential when operating AI. When considering a tool, choose one based not only on its domestic track record, but also on whether it has a rich international track record.

Step 5: AI learning period

To ensure that the AI learns efficiently, it is essential to work with an engineer or vendor with knowledge of data, statistics, and cloud computing.

Step 6: Customize AI tools

Cooperation with the vendor is essential to customize the tool to make it easy to use.

Step 7: PDCA Cycle

Set up a trial period for the AI and keep making improvements until a practical flow is established.

After the trial period, go into full-scale operation. Even after full-scale operation, periodically survey employees to review the AI workflow.

Final thoughts

The introduction of AI is progressing in India, China, and other countries around the world. We can expect AI to continue to be used increasingly in industries such as finance, healthcare, tourism, and telecommunications.

If you are unsure as to whether AI would be a worthy investment and are concerned with potentially wasting time and resources in the setup process, it may be a good idea to identify the specific efficiency issues facing your company. Consider whether you really need to introduce AI tools or whether RPA can replace them.

AI may be a difficult technology to understand, but it has many possibilities. Consider it not only as a tool to improve business operations, but also as a tool to start a new business.