We tend to discuss artificial intelligence in terms of capability: Which tasks will it accelerate? Which professions will it reshape? How quickly can it produce a text, an image or a piece of software?
I find another question more interesting:
When AI drives the cost of production towards zero, what becomes truly valuable?
One of the basic rules of economics is simple: when something becomes abundant, the scarce resources that complement it become more valuable.
Content, information, code, images and options are multiplying rapidly. Original perspective, critical thinking, trustworthiness and the ability to make sound decisions are not scaling at the same speed.
AI is changing not only how we produce, but also where value is created.
The human signature is becoming a signal of value
The “organic” label we see in supermarkets did not always exist. Once production methods changed and the natural option became the exception, we needed a way to identify it.
A similar distinction is emerging in creative work.
In 2025, the US-based Authors Guild launched its “Human Authored” certification programme. The mark indicates that a book’s literary expression was created by a person rather than generated by AI.
Research from Columbia Business School also suggests that people may assign less creativity and monetary value to the same work when they believe it was produced by AI. The presence of AI-generated work can, in turn, raise the perceived value of human-made work.
That does not mean everything made by a person will automatically be valuable. What will matter is the experience, intention, original point of view and accountability behind the work.
“A human made this” will not be a guarantee of quality. It will become a signal of meaning and responsibility.
As information becomes abundant, judgement becomes scarce
For a long time, expertise was largely associated with accumulating knowledge. Today, someone at the beginning of their career can use the right tools to approach the information available to an experienced specialist in seconds—at least in terms of the answer displayed on a screen.
But reaching an answer is not the same as knowing whether it is correct, applicable and trustworthy.
In a study by Microsoft Research and Carnegie Mellon University involving 319 knowledge workers, participants reported investing less critical-thinking effort as their confidence in AI increased. The study is based on self-reported behaviour and does not prove direct causation. It does, however, point to an important risk: the more useful AI becomes, the less inclined we may be to question it.
I have seen another version of this problem for years in travel technology. You can connect a hotel to dozens of sales channels. But if the mapping is wrong or the rates and availability are outdated, more connectivity does not create more value; it spreads the same error faster.
The difference between being connected and working correctly is much like the difference between AI producing an answer and a person arriving at the right decision.
The professionals who stand out in the future will not simply be those who know the most. They will be those who can ask the right question, recognise flawed data and reject an AI recommendation when necessary.
As access to information becomes more democratic, the centre of expertise is shifting from accumulating knowledge to exercising judgement.
Seeing is no longer enough to believe
In 2024, a finance employee in engineering company Arup’s Hong Kong office received a confidential transaction request that appeared to come from senior colleagues. Suspicious of the email, the employee joined a video call with people who looked and sounded like the executives and colleagues they knew. The meeting appeared genuine.
After the call, approximately $25 million was transferred to several accounts. It was later discovered that everyone on the call, except the employee, had been represented using fake video and audio. (The Arup case)
This incident does more than show how far deepfake technology has advanced. It exposes how strongly we still rely on the assumption: “I saw it, I heard it, I recognised the person.”
As synthetic images and voices become cheaper to produce, proving what is real will become more expensive. Trust will move from personal perception to verification processes. In a world where technology accelerates everything, one of the most valuable behaviours may be to pause and confirm a request through another channel.
Search is becoming recommendation
For years, one of the main goals of digital visibility has been to appear on the first page of Google. That still matters, but users now have a new intermediary in their decision journey.
People are increasingly asking AI to research options, compare them and recommend a small number that fit their needs.
According to Adobe’s US retail data, visitors referred by AI channels converted 42% better than visitors from other sources in March 2026. One year earlier, the same type of traffic converted 38% worse.
There is a natural connection here with travel technology and distribution. The purpose of distribution is to make the right product available at the right point of demand. AI interfaces are now becoming one of those demand points.
Instead of making a broad search for “hotels in Istanbul”, a traveller may soon ask an AI that knows their budget, purpose and preferences to recommend three suitable options.
At that point, simply being online will not be enough. A hotel or travel brand will need information that is current, consistent, understandable and trustworthy. Conflicting room details, missing content or a weak digital reputation may cause a brand to be misunderstood—or not recommended at all.
SEO is not disappearing; the meaning of visibility is expanding. The next competition is not only about ranking in search results. It is about being understood correctly by AI and recommended to the right customer.
The new price list
AI is not destroying value. It is moving it:
- As production becomes abundant, original perspective becomes more valuable.
- As information becomes abundant, judgement matters more.
- As synthetic media becomes easier to create, verifiable trust carries a premium.
- As options multiply, being recommended in the right context becomes more valuable.
I do not see this transformation as a decline in human value. On the contrary, AI is making the distinctly human contribution easier to see.
Producing something will matter less than understanding why it should be produced. Having an answer will matter less than recognising when that answer is wrong. Seeing will matter less than being able to verify.
AI can expand everyone’s capacity to produce. But the problem we choose, the decision we make and the signature we put on the outcome are still ours.
What is AI rapidly multiplying in your work—and which value that it cannot replicate is becoming more important?