“Cutting Staff with AI” Was a Fantasy
The latest survey reveals a shocking truth: even if you reduce staff by introducing AI, profits don’t increase. According to ASCII.jp, many companies aim for “workforce reduction” with AI, but this strategy may be fundamentally flawed.
From my own experience supporting AI adoption at over 38 clients, I can say that AI’s true value lies not in “replacing people” but in “augmenting people.” As someone who uses Claude Code and ChatGPT in daily operations, I’ve found that AI actually acts as a catalyst for creating new tasks.
The Core Issue Highlighted by SMEs Without IT Staff
Another key finding from the same survey: a staggering two-thirds of small and medium-sized enterprises (SMEs) have no dedicated IT staff. This isn’t just a staffing shortage.
The success of AI adoption hinges on post-implementation operational design. But in companies without IT staff, even basic questions like “Who will manage it?” and “What’s the security policy?” remain vague. As a result, many cases see AI tools underutilized, with no return on investment, left to gather dust.
In my consulting work, the first step isn’t tool selection but designing the “post-AI workflow.” For example, when automating contract review with AI, if you don’t decide beforehand “Who does the final check?” and “What’s the exception handling process?” you’ll end up with humans double-checking everything, and efficiency won’t improve.
The Accelerating Risk of User Account Leaks
Even more serious is the acceleration of user account leaks. Most AI tools are cloud-based, requiring account management. In companies without IT staff, risks like former employees’ accounts being left active or password reuse becoming the norm increase dramatically.
In fact, at one SME I supported, all employees were using the same password for ChatGPT. Before introducing AI, the first priority should be establishing basic account management. This is a responsibility business owners must take on themselves.
Three Ways to Boost Profits with AI
So, how can you truly increase profits with AI? Based on my practice, here are three effective methods.
1. Start by “Visualizing” Tacit Knowledge Tasks
The first step in AI adoption is eliminating task dependency on specific individuals. At my own company, before building pipelines for automated social media posts or WordPress article generation, we first visualized “who does what and how.” This clarifies which parts can be automated by AI and which require human judgment.
Specifically, follow these steps:
- All employees log their tasks for one week
- Categorize each task into “rule-based parts” and “judgment-required parts”
- Replace rule-based parts with AI
- Use the freed-up time for creating new value
2. Improve Accuracy by Using Multiple AIs
I use three AIs—Claude, ChatGPT, and Grok—depending on the situation. Relying on a single AI leads to biased answers and accuracy limits. Using multiple AIs enables cross-verification, significantly improving output quality.
In terms of cost, I can fully utilize all three AIs for about $140 per month. This investment generates roughly $50,000 worth of value annually. That’s an ROI of about 2,989%—a far more realistic option than cutting staff.
3. Break Down Data Silos and Build an AI Learning Foundation
The biggest factor affecting AI accuracy is the quality and quantity of training data. In many companies, data is siloed by department, preventing AI from accessing sufficient information.
From my experience, it’s crucial to first integrate data scattered across Google Drive, Slack, and various SaaS platforms, making it accessible to AI. This process can take weeks to months, but skipping it halves the effectiveness of AI adoption.
What SMEs Without IT Staff Can Do Right Now
Even companies without IT staff can start with these measures.
First, limit the scope of AI tool adoption. Instead of a company-wide rollout, start with a specific department or task—this is key to success. One of my clients began with AI for invoice processing in the accounting department and successfully expanded to other departments within three months.
Next, use external experts on a spot basis. Even if you can’t afford a full-time IT staff member, you can use AI consultants or IT support services for a few hundred dollars a month. I myself support AI adoption at multiple companies through two to three online meetings per month.
Finally, enforce basic security measures. Start with zero-cost steps like introducing password management tools or setting up two-factor authentication.
Summary: Rethink Workflow Design, Not Staff Cuts
Business leaders who aim for “workforce reduction” with AI are creating significant opportunity losses. AI’s true value lies in augmenting human capabilities and generating new value.
At my own company, we haven’t reduced staff since introducing AI. On the contrary, we’ve used the time freed up by AI to focus on developing new businesses and improving existing services. As a result, profits have been steadily increasing.
Isn’t the shift from “cutting staff with AI” to “empowering people with AI” the fundamental change required for future management?
Comments