A story about strategic ambiguity, intellectual evasion, and the power of answering a question without actually answering it.
October 2026. In the rapidly expanding technology district of Tokyo, a major artificial intelligence company called NexaCore was facing a crisis.
The company had recently launched an AI-powered platform designed to automate administrative work for hospitals, banks, and local governments. Its executives had promoted the technology as a revolutionary breakthrough, promising greater efficiency, lower operating costs, and fewer human errors.
But three months after its launch, serious problems began to emerge.
A software update had caused the system to incorrectly classify certain medical insurance applications. Hundreds of applications had been delayed, and several hospitals were struggling to reconcile their records. Although no major financial losses had yet been confirmed, the incident raised uncomfortable questions about the reliability of the company’s technology.
At 9:00 a.m. on Monday, an emergency meeting was convened.
Around the conference table sat the company’s CEO, senior engineers, legal advisers, and representatives from several affected hospitals.
At the head of the table was CEO Masato Kuroda, a 56-year-old executive renowned for his extraordinary intelligence and seemingly limitless ability to remain composed under pressure.
Across from him sat Naomi Fujita, a hospital administrator who had traveled from Osaka specifically to obtain a straightforward answer.
“Mr. Kuroda, we need to know one thing. Will your company compensate the hospitals for the additional costs caused by this software failure?”
The room fell silent.
Kuroda adjusted his glasses.
“That’s an extremely important question,” he began. “But before we discuss compensation, I think we need to reconsider the broader relationship between technological innovation and institutional responsibility.”
Naomi frowned.
“I’m asking whether you will pay for the losses.”
“Of course,” Kuroda replied smoothly, “and that’s precisely why we must avoid reducing this discussion to a simple financial transaction. Artificial intelligence is transforming the very definition of operational responsibility. In the past, responsibility was relatively easy to assign because decisions were made by identifiable individuals. Today, however, we operate within interconnected ecosystems involving developers, users, data providers, infrastructure operators, and regulatory authorities.”
He paused, allowing his words to settle.
“The real question, therefore, is not simply who pays, but how society should distribute responsibility in an increasingly automated world.”
The legal adviser beside him lowered his eyes, concealing a faint smile.
Naomi looked at him for several seconds.
“That’s an interesting argument. But will NexaCore compensate us?”
Kuroda nodded thoughtfully.
“Your concern highlights another important issue: the distinction between immediate remedies and long-term structural improvements. If we focus exclusively on compensation, we risk overlooking the systemic conditions that allowed this incident to occur in the first place.”
He then requested a presentation.
A large screen displayed a sophisticated diagram showing the relationship between AI model performance, data quality, human oversight, cybersecurity, and institutional governance.
Kuroda continued for another fifteen minutes.
He discussed machine-learning model drift, the importance of continuous monitoring, the limitations of automated decision-making, and the emerging international consensus around risk-based AI governance.
He even mentioned the European Union’s AI Act, whose risk-based framework distinguishes between different categories of AI applications and imposes obligations according to their potential impact. He explained that AI systems used in sensitive settings require particularly careful attention to transparency, human oversight, and accountability.
Everything he said was technically reasonable.
Everything sounded important.
And almost none of it answered Naomi’s question.
Finally, Naomi interrupted him.
“Mr. Kuroda, I understand the importance of AI governance. But our hospitals have already paid overtime to employees, hired temporary administrative staff, and spent additional money correcting the records. We need a commitment, not a lecture.”
For the first time, Kuroda’s expression changed slightly.
He folded his hands.
“Ms. Fujita, I appreciate the urgency. However, making a premature commitment could create expectations that might ultimately prove unsustainable. We must first establish a fair and scientifically defensible methodology for evaluating the impact of the incident.”
“So you’re not going to answer?”
“On the contrary. I believe I’ve explained why this issue deserves a much more comprehensive examination.”
The meeting continued for another forty minutes.
By the end, NexaCore had promised to establish a joint working group, commission an independent technical assessment, and prepare a report on the incident. There would be another meeting in three weeks.
But there was no compensation agreement, no payment schedule, and no commitment to cover the hospitals’ expenses.
As Naomi left the building, she turned to one of her colleagues.
“Did we get an answer?”
Her colleague shook his head.
“No. We got a presentation.”
Two days later, Kuroda received an internal memorandum from the company’s communications director.
It contained a transcript of the meeting and a proposed public statement:
“NexaCore remains firmly committed to working with all stakeholders to develop a sustainable framework for addressing the challenges associated with the integration of advanced artificial intelligence into essential public services.”
Kuroda read it, smiled, and approved it.
He knew that the statement was difficult to criticize. It expressed cooperation, responsibility, and concern for the future. Yet it contained no concrete admission, financial commitment, or deadline.
In negotiation theory, this kind of behavior can be understood as strategic ambiguity: maintaining room for maneuver by avoiding language that would bind the speaker to a particular outcome. In organizational communication, it can also resemble evasive answering or topic shifting—responding to a related but different question while creating the impression that the original question has been addressed.
There is an important distinction, however, between legitimate complexity and deliberate evasion. In a genuine technical investigation, it may be necessary to establish facts before assigning responsibility. But that does not prevent an organization from acknowledging the question, explaining what remains unknown, and setting a firm deadline for an answer.
Indeed, contemporary AI governance increasingly emphasizes precisely these issues.
For example, the European Union’s AI Act entered its main enforcement phase on August 2, 2026, although certain high-risk obligations have later application dates. Its framework emphasizes risk management, documentation, human oversight, and accountability throughout the lifecycle of AI systems. Shaping Europe’s digital future +1
Similarly, the United Kingdom’s October 6, 2026 government response accepted all 44 recommendations from its National Commission on AI healthcare regulation, including proposals to clarify responsibilities among manufacturers, healthcare providers, and other stakeholders. GOV.UK +1
These developments illustrate why sophisticated explanations alone are insufficient: organizations must ultimately translate principles into identifiable responsibilities and practical remedies.
Three weeks later, Naomi returned for the follow-up meeting.
This time, she brought a single sheet of paper.
At the top were three questions:
- Does NexaCore accept responsibility for the software malfunction?
- Will NexaCore compensate the affected hospitals?
- If the answer is not yet known, on what date will a final decision be made?
She placed the paper in front of Kuroda.
“I’ve simplified the discussion. You don’t need to explain the future of artificial intelligence. We just need answers to these three questions.”
Kuroda looked at the paper.
For a moment, he appeared genuinely uncomfortable.
Then he smiled.
“Ms. Fujita, I must commend you on the clarity of your approach. Simplicity is certainly an important virtue in complex organizational environments.”
Naomi stared at him.
“Please answer question one.”
And for the first time, the entire room understood that the most sophisticated answer could sometimes be nothing more than an elaborate way of saying nothing.
The ability to speak intelligently is not the same as the willingness to answer honestly.
Underlying themes and specialized concepts:
- Strategic ambiguity: Preserving flexibility by avoiding explicit commitments.
- Deflection: Redirecting a conversation toward a related but less uncomfortable subject.
- Impression management: Using sophisticated language to maintain an appearance of competence and responsibility.
- Responsibility diffusion: Allowing accountability to become fragmented among multiple departments or organizations.
- Procedural delay: Replacing an immediate decision with committees, assessments, and future meetings.
The central irony is that a person can say many intelligent and technically correct things while communicating almost nothing of practical value.
All names of people and organizations appearing in this story are pseudonyms
Comments