Introduction

In Japan, the shortage of IT engineers and system development professionals has become a serious problem.

The IT technologies that companies need, including web services, business systems, cloud infrastructure, AI, data analysis, and security, are becoming more sophisticated every year. Meanwhile, there are many companies that have been unable to secure sufficient talent who understand these technologies and can handle everything from design and development to operation.

In particular, Japan's working-age population is declining due to its low birthrate and aging population, and it is thought that solving the IT talent shortage in the future through the same recruitment methods and staffing arrangements as before will be difficult.

Therefore, what will become important for Japanese companies from now on is the adoption of AI (artificial intelligence).

I believe that AI should not be viewed merely as a convenient tool or as technology that assists with programming. AI should become an important management foundation that enables companies to perform more work with limited talent and maintain and improve productivity.

This article examines, from the perspective of system development, the current state of the engineer shortage in Japan, its background, how AI will change engineers' work, and ultimately why companies need to adopt AI.

1. The Current State of the Engineer Shortage in Japan

Japan's IT talent shortage is not merely a recruitment market issue. It is a structural challenge that also affects companies' DX initiatives and international competitiveness.

According to "DX Trends 2026," published by the Information and Processing Agency (IPA), an independent administrative agency, 85.5% of companies responded that they were "somewhat short" or "significantly short" in terms of the number of people needed to promote DX in its FY2025 survey.

Furthermore, 88.8% of companies feel there is a shortage in terms of the quality of talent.

In other words, this is not a problem that can be solved simply by increasing the number of hires; the professionals themselves, with the necessary technologies and specialized knowledge, are in short supply.

Source:
https://www.ipa.go.jp/digital/chousa/dx-trend/dx-trend-2026.html

These figures reveal the reality that Japanese companies, even if they wish to advance DX, cannot secure sufficient talent to make it happen.

For example, even if a company tries to introduce a new business system, it will need specialized professionals such as the following.

• System engineers who can organize business requirements
• Architects who can design entire systems
• Front-end and back-end developers
• Database and cloud specialists
• Security engineers
• AI and machine learning engineers
• Project managers
• System operations and maintenance personnel

These roles each require different specialized knowledge.

In particular, at many small and medium-sized companies, a small number of engineers often handle multiple roles.

A single engineer may be responsible for everything from defining requirements and designing systems to programming, testing, server setup, and operations and maintenance.

In such situations, the burden on each individual becomes greater, leading to slower development and making quality management more difficult.

2. Why is there a shortage of engineers in Japan?

The engineer shortage is caused by several overlapping factors.

2.1 Decline in the working-age population due to population decline and aging

One of the biggest factors is changes in Japan's demographic structure.

As the population declines and ages, the number of young people entering the labor market is decreasing.

This is not an issue confined to the IT industry, but it has particularly serious consequences in IT fields that require specialized knowledge.

IT technology is a field that requires continuous learning.

As programming languages, frameworks, cloud services, databases, and AI technologies continue to evolve, companies must continually secure talent who can adapt to changes in technology.

However, in an environment where the working population itself is shrinking, companies end up competing for a limited pool of engineers.

2.2 Increasing sophistication of IT technology and specialization of fields

In the early days of system development, there were cases where websites and business systems could be built using a relatively limited range of technologies.

However, the situation has changed dramatically in today's development environment.

For example, even a typical web system may combine technologies such as the following:

• Front-end technologies such as React and Next.js
• Back-end technologies such as Node.js, Python, and PHP
• Databases such as PostgreSQL and MySQL
• Cloud environments such as AWS and Google Cloud
• Development and operations infrastructure such as Docker and CI/CD
• API integration and integration with external services
• Authentication, access control, and data protection
• Artificial intelligence technologies such as AI, LLMs, and RAG

It is now an era in which simply being able to write programs is no longer enough to build a sufficient system.

The ability to understand the architecture of the entire system and appropriately combine technologies is required.

In other words, as IT demand increases, the level of expertise required of engineers is rising as well.

2.3 Corporate hiring and development are not keeping up with demand

Another challenge is companies’ talent development systems.

Companies tend to seek experienced engineers who can contribute immediately.

However, experienced talent is limited in the market, and competition to hire them has intensified.

Meanwhile, even when novice or young engineers are hired, they require training time and guidance from instructors before they can excel in practical work.

The problem here is that those responsible for training also lack sufficient capacity.

When existing engineers are busy with routine work and development projects, they cannot secure the time to train junior staff.

As a result, a vicious cycle emerges: companies want to hire experienced people but cannot, and want to develop junior engineers but have no capacity to train them.

2.4 Maintaining legacy systems also requires talent

Many Japanese companies also have old business systems that have been used for many years.

These systems may use programming languages that are no longer common today, as well as older frameworks.

For example, systems that use older versions of PHP and Java, as well as COBOL and CakePHP.

We need not only engineers who can adapt to new technologies, but also people who can understand the specifications of existing systems and carry out troubleshooting and upgrades.

If technical staff who understand an old system leave the company, even maintaining that system may become difficult.

Thus, the shortage of engineers in Japan affects not only the people needed to develop new systems, but also those needed to maintain existing IT assets.

3. How will the advancement of AI change the work of engineers?

The presence of AI is important here.

The emergence of generative AIs such as ChatGPT, Claude, and Gemini has dramatically changed the way software is developed.

Today’s AI does more than simply answer questions.

It can support a wide range of development tasks, including code generation, program correction, error analysis, test code creation, technical research, documentation, and data analysis.

For example, even for errors that engineers would previously have spent considerable time investigating, presenting the logs and related code to AI enables possible causes and fixes to be organized in a short time.

It is also possible to create an initial program implementation from a specification, or to analyze existing code and suggest improvements.

Such technology may improve productivity in development environments.

3.1 Development tasks that can be automated and made more efficient by AI

Specifically, AI can make the following tasks more efficient.

Development Process	AI Assistance
Requirements Definition	Classification of requirements and drafting of specifications
Basic Design	Proposing data structures and system architecture
Detailed Design	Organizing API specifications and processing flows
Programming	Code generation and refactoring
Testing	Creating test cases and test code
Debugging	Analyzing the causes of errors and presenting proposed fixes
Documentation	Creating technical documents and operation manuals
Operation and Maintenance	Supporting log analysis and incident investigation

By combining these tasks with AI, engineers can reduce the time spent on routine work and focus more on important design and decision-making.

However, this does not mean that code generated by AI can be deployed to the production environment as is.

In practice, humans need to verify consistency with the specifications, security, performance, maintainability, and other aspects.

AI should be used not as an entity that assumes all responsibility on behalf of developers, but as something that expands developers' capabilities.

3.2 Will AI Make Engineers Unnecessary?

As AI becomes more widespread, there are opinions that “programmers and engineers may no longer be needed in the future.”

However, I do not think so.

This is because even if AI automates some tasks, system development itself does not disappear.

For example, even if a customer wants to build “a system that streamlines business operations,” those words alone are not enough to build a concrete system.

In practice, the following judgments are required.

• What is the problem the customer truly wants to solve?
• What challenges exist in the current business workflow?
• Which features should be developed as a priority?
• What data structure would be appropriate?
• How should security and personal information be protected?
• How should future feature expansions be accommodated?
• How should AI responses and generated results be verified?

These issues cannot be resolved through simple code generation alone.

Therefore, going forward, I believe engineers will need not only the ability to write programs, but also the ability to understand business operations, analyze problems, and design appropriate systems using AI.

4. Why Japanese Companies Must Introduce AI

This is the point we most want to convey in this article.

For Japanese companies, introducing AI is not merely a response to a technological trend.

Given the future decline in the labor force and the shortage of IT professionals, it is an important option for continuing business and maintaining competitiveness with limited human resources.

4.1 The Need to Supplement Talent Shortages with AI

In conventional corporate management, it was common to increase staffing as the volume of work grew.

However, an increasing number of companies may be unable to hire all the talent they need in the future.

Companies therefore need to create systems that enable existing employees to handle more work, rather than relying solely on adding staff.

Possible applications include the following:

• AI-powered customer inquiry handling
• Searching internal FAQs and manuals
• Extracting data from invoices and other documents
• Creating routine emails and reports
• Classifying and analyzing customer information
• Recording and organizing sales activities
• Automating reservation and reception operations
• Improving the efficiency of system development and testing

If these are implemented effectively, companies can reduce repetitive work performed by employees and allow them to devote more time to higher-value-added activities.

AI is expected to become an important means of supporting business operations, particularly for small and medium-sized enterprises that have difficulty establishing dedicated IT departments.

4.2 The Productivity Gap Between Companies That Introduce AI and Those That Do Not

In the future, one of the factors determining competitiveness between companies is likely to be the extent to which they can integrate AI into day-to-day operations.

Even among companies with the same number of employees, one may manually organize information, prepare materials, and respond to inquiries, while another uses AI to streamline operations.

In the latter case, the company may be able to handle more work with the same time and number of employees.

Of course, simply introducing AI does not automatically improve productivity.

What is important is to clearly identify which parts of the business to apply AI to and to incorporate it into actual business processes.

According to the IPA's “DX Trends 2026,” 58.0% of companies have introduced or are testing AI.

The proportion of companies introducing generative AI has also increased from 22.6% in fiscal 2024 to 44.0% in fiscal 2025.

However, the proportion of companies that have integrated generative AI into their operational business processes remains at only 11.9%.

In other words, an increasing number of companies are using AI, but there is still much room to go from the stage of individual use to the stage of transforming work across the entire organization.

I believe this gap will determine companies’ future competitiveness.

Merely distributing AI accounts to employees is not enough.

It is important to integrate AI with business systems, databases, customer management systems, internal workflows, and other systems, enabling it to be used continuously in day-to-day operations.

4.3 Not adopting AI itself becomes a business risk

When adopting AI, it is necessary to consider costs, security, and other factors.

However, there are also risks associated with not adopting AI.

For example, while competitors are improving operational efficiency with AI and accelerating the delivery of services, what would happen if only your company continued doing the same work manually as before?

Operations could stall because of labor shortages, responses to customers could be delayed, and development costs could increase.

Furthermore, limited engineering resources could become consumed by routine maintenance work, leaving engineers unable to focus on developing new services and businesses.

To avoid such situations, companies need to advance the adoption and use of AI in stages rather than postponing its introduction.

The key here is not to replace all work with AI.

Rather, it is to identify work that requires human judgment and work that can be entrusted to AI, and to design their respective roles appropriately.

5. Skills engineers will need in the AI era

As AI becomes more widespread, the skills required of engineers will change as well.

Previously, it was important to be proficient in specific programming languages and frameworks.

Of course, this technical knowledge will continue to be necessary.

However, once AI can assist with code generation and technical research, an engineer’s value can no longer be judged solely by how quickly they can write code.

The abilities considered particularly important going forward are as follows.

5.1 The ability to use AI skillfully

This is the ability to give AI appropriate instructions, provide it with the necessary information, and evaluate the results it generates.

It is important not merely to enter prompts, but to use different approaches with AI according to the objective.

For example, AI can be used in each stage of a process, including code generation, design reviews, technical research, and test creation.

5.2 The ability to design entire systems

Even if AI can generate code, a company cannot provide stable services over the long term if the design of the entire system is inappropriate.

It is important to have the ability to understand the entire system, including database design, API design, authentication, authorization management, infrastructure configuration, and contingency measures.

5.3 Business Knowledge and Problem-Solving Skills

Engineers must understand the problems customers are facing, rather than merely implementing the features they have been asked to build.

For example, when systematizing operations that are currently managed manually, they must analyze the existing workflow and determine what should be automated.

AI can assist with that analysis, but business priorities and final decisions must remain the responsibility of humans.

5.4 Security and Quality Management Skills

AI-generated code may contain errors or vulnerabilities.

Therefore, knowledge of code review, testing, vulnerability assessment, data protection, and other areas is becoming increasingly important.

Precisely because they use AI, companies need engineers who can correctly evaluate the generated results.

5.5 Ability to Continuously Learn

AI technology is evolving at an extremely rapid pace.

Therefore, engineers need to continually learn new technologies and development methodologies rather than depending on specific tools alone.

Future engineers will require both programming skills and the ability to effectively use AI.

6. How Should Companies Introduce AI?

Many companies probably understand the need for AI but do not know how to introduce it in practice.

I do not believe it is necessary to develop a large-scale AI system from the outset.

What is important is to introduce AI on a small scale in line with the company’s challenges and expand it while verifying the results.

Specifically, I recommend proceeding in stages such as the following.

STEP 1: Analyze Current Operations

First, identify work within the company that takes a considerable amount of time or occurs repeatedly.

Examples include handling inquiries, entering data, creating documents, searching for information, and managing customers.

STEP 2: Select Operations That Can Be Made More Efficient with AI

AI is not suitable for every operation.

You should begin with areas where AI is expected to be effective, such as routine tasks and work involving organizing large amounts of information.

STEP 3: Select the Appropriate AI Technology

Depending on the objective, you can use generative AI such as ChatGPT, Claude, or Gemini.

You can also combine API integration, RAG, AI agents, OCR, automation tools, and other technologies according to the operation.

There is no need to develop an AI model exclusively for your own company from the outset.

By leveraging existing AI services, you may be able to reduce implementation costs and development time.

STEP 4: Integrate with existing systems

It is important not only to use AI independently, but also to integrate it with existing business systems and databases.

For example, integrating AI with a customer management system may improve the efficiency of searching customer information and handling inquiries.

Also, enabling internal documents to be searched using RAG can reduce the time employees spend finding the information they need.

STEP 5: Measure results and continuously improve

After introducing AI, check working hours, processing volume, error rates, costs, customer satisfaction, and other metrics.

Comparing conditions before and after introduction allows you to determine whether the expected effects are being achieved.

At the same time, security measures are also necessary, including the handling of personal and confidential information, incorrect AI responses, access permissions, and log management.

Introducing AI is not a one-time endeavor that ends upon implementation. Continuous improvement is important.

7. Can AI solve the shortage of engineers?

From what has been discussed so far, it is clear that AI will not completely eliminate the shortage of engineers.

AI also has its limitations.

For example, understanding complex business requirements, organizing ambiguous specifications, designing entire systems responsibly, and making security-related judgments still require advanced expertise.

Furthermore, using AI effectively also requires people who understand how it works and the risks involved.

In fact, in IPA's “DX Trends 2026,” the proportion of companies identifying a “shortage of specialized professionals” as a challenge in AI implementation and operation reached 50.1%.

Moreover, labor shortages continue across many occupations related to AI.

In other words, introducing AI does not make human talent unnecessary; it creates a new need for people who can properly implement and operate AI.

I believe that, in the future of the IT industry, differences will emerge between conventional engineers and engineers who can leverage AI in terms of productivity and the range of work they can handle.

The key is not simply to increase the number of engineers, but to create a development environment in which each individual can leverage AI to produce greater results.

8. What Japanese companies need going forward is “collaboration between humans and AI”

What Japanese companies should strive for is a development and operational system in which humans and AI collaborate, each leveraging their own strengths.

Humans have strengths such as creativity, business understanding, experience-based judgment, accountable decision-making, and communication with customers.

AI, on the other hand, has strengths in processing large amounts of information, generating text and code, analyzing data, and improving the efficiency of repetitive tasks.

By combining the two effectively, it becomes possible to handle more work with fewer personnel.

For example, even in a development project that previously required five engineers, leveraging AI for design assistance, code generation, and test automation may reduce the time required to complete the work.

However, this does not necessarily mean reducing five people to one.

The number of people required varies depending on the scale and complexity of the project and the quality requirements.

Rather, it is important to use the time saved through AI to develop new features, improve quality, strengthen security, and create customer value.

Also, to extend the efficiency gains from AI across the entire company, it is necessary to create an environment in which employees in every department—not just engineers, but also those in sales, accounting, general affairs, and marketing—can use AI effectively.

AI is not a technology for the IT department alone.

We believe that its value is maximized when it is leveraged across the company as a whole.

9. Summary: AI adoption will become an inevitable direction for Japanese companies

The shortage of engineers in Japan is a challenge that will have a major impact on future business activities.

Multiple problems are overlapping, including the shrinking labor force resulting from declining birthrates and population aging, advancements in IT technology, growing demand for DX, and insufficient cultivation of specialized talent.

If companies continue to rely solely on recruiting talent, as they have traditionally done, it may become difficult to maintain the development capacity and business-processing capabilities that they require.

Therefore, the introduction and use of AI will become increasingly important going forward.

However, introducing AI itself is not the goal.

What is truly important is to improve how work is organized by leveraging AI and create an environment in which people can focus on more important work.

I believe that, for Japanese companies going forward, AI adoption will not be merely one option, but an important management strategy for maintaining productivity and competitiveness in a society with a shrinking labor force.

And for engineers as well, it is important not to avoid AI, but to understand it, master its use, and acquire the ability to integrate it into appropriate systems.

Going forward, companies will need engineers who can not only write programs, but also use AI to solve problems, design entire systems, and provide new value to the company.

Leverage AI to address the shortage of engineers. And train engineers to leverage AI.

I believe that advancing both of these efforts simultaneously will be necessary for Japan's IT industry and companies to overcome the challenges of the times ahead.

Thank you very much for reading to the end.