Artificial intelligence is entering almost every serious business discussion. Some people see it as a revolution. Others see it as a threat. I see it as a powerful business instrument—but an instrument must still be guided by someone who understands the purpose, the consequences and the people involved.
After more than two decades in business, I have learned that technology can improve speed, but speed alone does not create a successful institution. A business also requires judgment, accountability, relationships and the courage to make difficult decisions. These qualities cannot simply be downloaded.
AI can accelerate work, but it cannot carry responsibility
AI can analyse information, organise documents, identify patterns, improve communication and reduce repetitive work. Used properly, it can save valuable time and help teams operate more efficiently.
However, efficiency and leadership are not the same thing.
When an important decision affects an employee, customer, investor or community, someone must accept responsibility for that decision. A computer system cannot stand before an unhappy customer, protect an organisation’s reputation or explain why a particular choice was morally and commercially correct.
The final responsibility will always belong to a human being.
That is why I do not believe leaders should ask, “Can AI make this decision?” The better question is, “How can AI help me make a better-informed decision?”
The transformation is real
The debate is no longer about whether AI will influence business. It already is.
The International Monetary Fund estimates that almost 40% of employment worldwide is exposed to AI in some form. The World Economic Forum reports that AI and information-processing technologies are among the developments most employers expect to transform their businesses before 2030.
These findings should be taken seriously, but they should not be interpreted as predictions that every exposed job will disappear. Exposure can also mean that a person’s work becomes faster, more accurate or more valuable with the assistance of technology.
In my view, the greatest risk is not that every human being will be replaced. The greater risk is that professionals and organisations that refuse to learn will gradually become less competitive.
Automation without judgment creates new risks
Every technology brings advantages as well as fresh responsibilities.
AI can produce an answer confidently even when the underlying information is incomplete or incorrect. It can repeat bias contained in its source material. Sensitive business information can also be compromised if employees use public tools without understanding confidentiality and data-protection requirements.
A polished document is not automatically an accurate document. A quick answer is not necessarily the correct answer.
This is where experienced leadership becomes even more important. Leaders must decide what information can be trusted, what must be verified and what should never be delegated to a machine.
Legal decisions require qualified legal review. Medical decisions require medical professionals. Financial decisions require proper analysis and accountability. Technology can assist these professionals, but it should not be used to bypass them.
The real advantage will come from people who learn to work with AI
I believe the future will not be divided simply between humans and machines. It will be divided between people who know how to use technology responsibly and those who do not.
A capable employee using AI intelligently may complete certain tasks faster, examine more information and communicate more clearly. But the employee must still understand the subject. Without knowledge, it becomes difficult to recognise when the technology is wrong.
Therefore, businesses should not treat AI merely as a cost-cutting mechanism. They should treat it as a skill-development opportunity.
Training people is more valuable than frightening them. Employees should learn how to frame questions, verify results, protect confidential information and apply their own professional judgment. Organisations that combine experienced people with modern tools will be stronger than organisations that depend entirely on either one.
Five principles I would follow
1. Use AI to assist, not to avoid accountability
Every important output should have a responsible human owner. Someone must verify it and be prepared to stand behind it.
2. Never compromise confidentiality
Customer information, legal documents, financial records and internal business strategy must be handled carefully. Convenience should never become an excuse for careless data sharing.
3. Verify important facts
AI-generated information should be treated as a starting point for research—not as unquestionable evidence. Important claims should be checked against reliable and, wherever possible, primary sources.
4. Invest in employee training
A company cannot introduce new technology and assume everyone will automatically understand how to use it responsibly. Training must cover practical use, limitations, ethics and data security.
5. Preserve the human relationship
Businesses are ultimately built between people. A customer remembers how an organisation behaved when a problem occurred. An employee remembers whether management treated them fairly. Technology should improve these relationships, not make the organisation feel distant and mechanical.
Leadership will become more important, not less
Some people assume that better technology will reduce the need for leadership. I believe the opposite is true.
When information becomes abundant and decisions become faster, organisations require even greater clarity. Leaders must determine what deserves attention, which risks are acceptable and where human values must take priority over convenience.
AI may help prepare a proposal, but it cannot build trust with a handshake. It may analyse a disagreement, but it cannot repair a damaged relationship through sincerity. It may calculate the most efficient option, but it cannot decide what kind of institution we want to build.
Technology will continue to evolve. Businesses must evolve with it. But progress should not mean surrendering our judgment.
The strongest organisations of the future will not be those that use the most AI. They will be those that use it with the most intelligence, discipline and humanity.
Sources and context
The employment and business-transformation figures above are research estimates and employer expectations, not guaranteed outcomes. See the International Monetary Fund’s analysis and the World Economic Forum’s Future of Jobs Report 2025.
— Tushar Kumar