Conclusion

AI is “more trusted” than humans.

An AI chatbot can build a relationship of trust in just a few days, a process that takes a human salesperson six months.

This is not a story about the future. A collaborative research team from Amrita University in India, the University of Venice, the University of Melbourne, and Ben-Gurion University in Negev compared AI chatbots and humans in a “trust-building showdown.” In terms of the ability to gain trust through long-term conversations, AI surpassed humans.

What this fact implies is two completely opposite futures:

• A future where the cost of building trust for small businesses approaches zero
• A future where the cost of building trust for scammers also approaches zero, leading to an infinite inflation of the cost of “not being deceived”

The same technology can be a weapon for both salespeople and scammers. It is meaningless to say, “Let’s streamline customer interactions with AI” without understanding this structure.

—

Why Can AI Earn Trust More Than Humans?

The reason is simple. AI lacks “fatigue,” “emotional fluctuations,” and “variability in responses.”

A human salesperson’s quality of response can change from Monday morning to Friday evening. There are days when they are in a bad mood. If there are ten customers, it is impossible to provide the same quality of response to all ten.

AI chatbots are different. They respond 24/7, maintaining the same tone while perfectly remembering the past statements and preferences of the other party. The moment a human feels, “This person remembers me well,” is the starting point of trust, and AI can execute this with 100% reproducibility.

Moreover, AI can be optimized to “convey what the other party wants to hear, at the right time, and in a way that is easy for them to accept.” This technique is called “building rapport” in the sales world, but AI can execute this in a data-driven manner, simultaneously for thousands of people.

A human salesperson can build a deep relationship with at most 3 to 5 people in a day. AI has no upper limit.

—

What Happens When Scammers Get This Weapon?

This is the flip side of the main topic.

The research team focused on a scam technique called “pig butchering.” This involves building a relationship of trust over weeks to months on social media or dating apps, ultimately leading to investment fraud. The damage amounts to hundreds of thousands to millions of dollars per case. According to an FBI report in 2023, the total damage from this technique reached approximately $4 billion (around 600 billion yen) annually.

Until now, this scam required “human scammers.” The number of targets a single scammer could handle at the same time was limited. In other words, there was a bottleneck of “labor costs” in scams.

AI eliminates this bottleneck.

If a single scammer uses an AI chatbot, they can automatically build “trust” with hundreds or thousands of people simultaneously, and do it better than humans. In fact, OpenAI reported in 2024 that a Cambodian scam organization was misusing ChatGPT and subsequently suspended their account. This is just the tip of the iceberg.

In other words:

The cost of “creating trust” has plummeted. At the same time, the cost of “discerning trust” has skyrocketed.

—

What Changes for Small Businesses — Thinking in Numbers

Let’s break this down in the context of local small businesses.

Traditional Trust-Building Costs

• Annual labor cost for one salesperson: 4 to 6 million yen
• Time taken for a newcomer to develop into a “trust-building salesperson”: 1 to 3 years
• Training and OJT costs during that period: 500,000 to 1 million yen annually
• Introduction and operation of customer response systems (CRM, etc.): 300,000 to 2 million yen annually
• Number of customers a single salesperson can maintain deep relationships with: 50 to 100 companies

If two salespeople are assigned to maintain relationships with 100 customers, the annual cost would be at least 9 million to 14 million yen.

After Implementing AI Chatbots

• Chatbot construction and operation costs: 30,000 to 100,000 yen per month (360,000 to 1.2 million yen annually)
• Number of customers that can be handled: effectively unlimited
• Operating hours: 24/7
• Variability in response quality: zero
• Setup period: a few days to a few weeks

Annual costs could potentially drop to less than one-tenth of the traditional costs. A world where 12 million yen becomes 1 million yen.

Moreover, what’s crucial is that this is not just about “cost reduction.”

The biggest challenge for small businesses is “person dependency.” When a star salesperson leaves, the relationship with customers disappears. AI chatbots do not leave. They continue to respond while retaining the entire conversation history with customers. The elimination of person dependency and the systematization of trust-building can be achieved simultaneously.

—

However, “Inflation of Trust” Will Occur

Here, we need to pause and think.

The decrease in the cost of trust-building with AI means that everyone has the same weapon. Your company, your competitors, and even scammers.

Consider it from the consumer’s perspective. Among the messages received daily, which are from genuine companies and which are from AI scammers? It will become indistinguishable.

What happens then?

The value of “trust built by AI” decreases.

This is inflation of trust. As AI becomes easier to “manufacture” trust, consumers will begin to discount the value of “trust via AI.” As a result, what can truly differentiate is “trust that cannot be replaced by AI” — that is, real connections, local achievements, and visible relationships.

Here lies the structure for small businesses to turn the tables.

—

A Structure Where Small Businesses Can Win

Large corporations can deploy AI chatbots on a large scale. However, large corporations lack “visible relationships.” When an AI bot from a Tokyo headquarters tells local customers to “trust us,” there are limits.

Small businesses, especially local ones, are different.

• The face of the president is visible
• They can meet at local events
• They can make direct phone calls if something happens
• Their local reputation is visible

Building AI chatbots on top of this “real trust foundation” is the right way for small businesses to compete.

Specifically:

1. Delegate initial contacts and daily interactions to AI — inquiry responses, FAQs, first responses to quote requests, follow-up messages
2. Humans step in during critical moments — handling complaints, negotiating large deals, expressing gratitude to long-term customers
3. Feed AI conversation data back to human sales — information such as “This customer is recently interested in XX” or “There are signs of dissatisfaction with YY” can be utilized by humans for the next action.

AI handles “quantity,” while humans handle “quality.” This combination is difficult for large corporations to replicate. This is because the “human quality” of large corporations is standardized manual responses, fundamentally different from the “context-aware responses” that small business owners and veteran employees possess.

—

So, What Should We Do?

Just do three things.

1. First, implement a cheap chatbot and let it handle part of customer interactions.

There are many services that can be started for a few tens of thousands of yen per month. There’s no need to aim for perfection. Just being able to automatically respond to 70% of frequently asked questions can dramatically change the burden on the ground.

2. Decide on “things that AI will never do.”

Clearly establish rules for handling complaints, apologies, and year-end greetings to important customers — these should be done by humans. The delineation between what AI handles and what humans handle determines the quality of trust.

3. Simultaneously establish a system to “not be deceived by AI” in your own company.

AI scam emails disguised as business partners or customers will surely increase in the future. A simple training session for employees (30 minutes is enough) and a review of confirmation flows during payments and contracts should be done immediately. Both offense and defense should progress simultaneously.

—

The Essence of This Discussion

We are entering an era where the cost of “creating trust” approaches zero. This is a massive opportunity for small businesses. Customer interactions that were previously only possible for large corporations can now be achieved for a few tens of thousands of yen per month.

However, at the same time, as trust can be “cheaply and abundantly manufactured,” the value of trust itself changes. Trust created by AI will eventually become commoditized.

The last remaining differentiating factor will be the real trust that says, “I trust this person” or “I rely on this company.” That is something that local small businesses inherently possess.

Optimize where efficiency can be achieved with AI. But never cut corners where only humans can do the work. Companies that can make this distinction will survive in the next decade.