Jobs Taken by AI vs. Jobs That Are Not: A Guide to Categorizing Consulting Tasks

Introduction

Hello, I am Ohno, the representative.

I am starting a new column series on the theme of survival strategies for consultants in the AI era.

In this series, I will explain in detail the career strategies for consultants to survive the AI era through practical themes such as "What is the unique value of humans that cannot be replaced by AI?", "Why is changing jobs from an engineer to an IT consultant the optimal solution right now?", and "The new era of self-contained consulting as a way of working."

In this first installment, I will share my thoughts on a specific method for categorizing "task replacement by AI," which is a topic many of you are concerned about.

As the debate over "AI taking consulting jobs" intensifies, many consultants are feeling anxious. However, not all consulting tasks will be replaced by AI. The important thing is to accurately understand which tasks are easily replaced by AI and which tasks can only be performed by humans.

If you can make this distinction, you will be able to use AI as a powerful weapon rather than a threat, and focus on high-value tasks that only humans can perform.

It is my belief that while "tasks with little value for humans to perform" in consulting work may be rapidly replaced, "areas that only humans can handle" will become more valuable than ever before.

Characteristics of tasks easily taken by AI

With the advent of generative AI and large language models, it is already possible to obtain a certain level of output in the initial phases of large-scale industry research and analysis. In particular, parts of intellectual work such as "repetitive information gathering and organization" are expected to become easier to delegate to AI in the future.

Basic research that used to take several days, such as industry trend surveys, collecting basic information on competitors, gathering market size data, and researching regulations and legal systems, can now be completed in a few hours using generative AI. Research based on publicly available information is a field where AI particularly excels.

In addition, AI can create high-quality, template-based outputs in a short time, such as creating the base for PowerPoint presentations, preparing routine analysis reports, visualizing data, and creating and organizing meeting minutes.

Furthermore, processing and analyzing numerical data—such as basic financial calculations, KPI analysis and trend identification, benchmark analysis, and initial processing of quantitative analysis—is an area where AI excels the most. It can process complex formulas instantly and execute analysis based on pattern recognition with high precision.

The essence of tasks that will not be taken by AI

On the other hand, the essential strengths in consulting can be organized from three major perspectives.

First, there is the task of eliciting a sense of conviction in decision-making and leading consensus-building.

In strategy formulation, AI has become capable of presenting optimal options based on vast amounts of data. However, corporate decision-making is not just about logical correctness; the sense of conviction among stakeholders is crucial.

The process of verbalizing the anxieties and dilemmas faced by management, appropriately organizing the pros and cons, and guiding the client to be convinced that "this is the decision to make" is something only humans can do.

AI can present options, but it is the consultant's role to judge "which option is truly optimal" and to support the decision-making process.

Second is the task of bridging departments and ensuring penetration into the field.

No matter how excellent a strategy is, it will not be executed if it is not accepted by the front lines. Especially in large companies, vertical structures and organizational silos are prevalent, and measures may not be implemented or may end up half-finished due to resistance from the field.

While AI can suggest 'what should be done,' it cannot handle the coordination required to 'make the field move.'

It requires 'translating strategies into a form that can be executed on the front lines and breaking through internal barriers,' such as involving frontline leaders, laying the groundwork, coordinating the interests of various departments to build consensus, and sometimes setting up forums for dialogue between management and the field.

Third is long-term accompaniment to ensure that transformation takes root.

Even if measures are introduced to the field, it is not rare for things to return to their original state over time.

Especially in organizational transformation or business reform, cases where 'there was enthusiasm at first, but before we knew it, things returned to the old way' occur frequently. AI can plan 'how to proceed with a project,' but it is not good at 'changing the consciousness and behavior of stakeholders over the long term.'

Ensuring the sustainability of organizational change and guiding clients to a state where they can operate independently is also a value that cannot be replaced by AI.

Practical Criteria for Sorting

When actually sorting these tasks, it is appropriate and effective to think using three criteria.

First, judge by 'routineness.'

Tasks with fixed procedures, tasks with predetermined output formats, and tasks that involve repeating the same work tend to be easily AI-driven.

On the other hand, tasks that require a different approach each time, tasks that require creativity or originality, and tasks that require customization to the other party are areas that should be handled by humans.

Next, judge by 'emotional elements.'

Tasks centered on logical analysis, where numerical data is the main basis for judgment and an objective correct answer exists, are easily AI-driven, whereas tasks that require consideration of the other party's emotions and values, tasks where a sense of conviction and empathy are important factors, and tasks that require building relationships of trust are areas that should be handled by humans.

Finally, judge by 'complexity.'

Tasks where variables are limited and can be patterned, and which can be learned from past cases, are easily AI-driven, but tasks where many variables are intricately intertwined, there are many unpredictable elements, and human intuition and experience are important are areas that should be handled by humans.

The direction consultants should aim for in the AI era

With the rise of AI, some consulting tasks will certainly be automated. However, this is an opportunity, not a threat. By delegating repetitive and routine tasks to AI, consultants can focus on their true value: the support that only humans can provide.

An optimal solution that ignores factors like internal power dynamics, interdepartmental friction, or organizational culture will never be implemented, no matter how excellent the proposal is. What is required here is not just analytical ability, but the power to deeply understand the client's situation and translate strategies into a feasible form.

Such 'human-centric perspectives and coordination skills' will continue to be needed as the essential value of consulting.

The key is to accurately sort your tasks, view AI as a tool, and continue to refine the value that only humans can provide while working in collaboration with it.

Consultants who can perform this sorting will be the ones truly needed in the AI era.

Conclusion

If you would like to discuss your career as a consultant in the AI era in more detail, please feel free to sign up for a career consultation through our recruitment service, talentstar.

For those considering a career change from engineer to IT consultant, we will propose the optimal career path tailored to each individual's skills and aspirations.

We are currently running a book giveaway campaign. Those who sign up for a consultation will receive a free copy of the book (IT Consultants in the AI Era). We hope you will take this opportunity to think about the possibilities of your career with us.

Next time, we plan to explain in detail: 'Research is 10 times faster with Generative AI: But why that alone isn't enough to make a living.' We will look specifically at the efficiency of AI utilization and the human value that remains regardless.

We look forward to helping you with your career development.