That uncle is now ChatGPT

Twenty years ago if you needed a good CA, a contractor who would actually turn up or the best kaju katli in the area, you did not search for anything. You asked an uncle. He had one name, he said it with complete confidence and you went there. Nobody went back for a second option, because the asking was the whole research process and it was done.

That uncle is gone. Something else now sits in his chair and it has read more than every uncle who ever lived put together. Ask ChatGPT, Gemini or Perplexity for the best packaging supplier in Pune or an interior contractor in Andheri and it answers exactly the way he did: three names, total confidence, no doubt anywhere in the sentence.

We are living through one of those rare moments when the world ends without announcing itself and a new one starts while everybody is still busy with their day. For the first time, the voice a buyer trusts before deciding is not a person. Humans and machines now decide together who gets chosen. The world you built your business in is not coming back.

That sounds frightening. It is supposed to, because every new world has frightened the people standing at its edge. This paper is about why that fear is the most predictable thing in human history, why it always passes and what does not change on the other side.

0101The new world, in numbers

In August 2026 an Indian company called DareAISearch published the SearchScore.AI Discovery Index. It ran 59,620 prompts through ChatGPT, Gemini and Perplexity, covering 2,981 brands across more than fifteen industries and scored how strongly each brand appeared when people asked for a recommendation. Only 19.3% reached high visibility. The other 80.7% fell below the line and the average score was 30 out of 100. (Figure 1)

The figure is the company’s own benchmark, but you do not need it to see the shift. Type your own category into one of these tools tonight.. a handful of names come back and many businesses that have done good work for forty years are simply not there.

0202Why we believe the machine

You would expect people to be wary of a machine telling them where to spend their money. The research says the opposite.

Jennifer Logg, Julia Minson and Don Moore (2019, Organizational Behavior and Human Decision Processes) asked people to make estimates, such as a person’s weight from a photo or how a song would rank, then gave them advice. The advice was identical, only the label changed: some were told it came from an algorithm and some that it came from other people. People moved further towards the advice when it was labelled as an algorithm. The authors called it algorithm appreciation.

Herbert Simon (1956, Psychological Review) explains why that matters so much. People rarely search for the best option, they search until something is good enough and then they stop. He called it satisficing. An AI answer is built for exactly that kind of stopping. It gives three names, it sounds certain and the buyer who already trusts it a little more than a person has no reason to look for a fourth.

So when an AI names three brands in your category, the question is not whether you are fourth. For that buyer there is no fourth.

0303Every new world arrives with a panic

Around 370 BC Plato wrote a dialogue called the Phaedrus, in which Socrates tells the story of an Egyptian king called Thamus. The god Theuth brings him the gift of writing and promises it will make people wiser. The king is not impressed. Writing, he says, will make people forgetful, because they will trust marks on a page instead of their own memory and they will seem wise without being wise. Socrates does not stop at the story either, he adds his own complaint that a written word cannot answer a question or defend itself. So the fear belonged to both of them and it is written down, which is the only reason we know they ever had it.

In 1545 a Swiss scholar called Conrad Gessner warned that the printing press had unleashed a “confusing and harmful abundance of books”. In late eighteenth century Germany doctors and clergy wrote anxiously about Lesesucht, a reading addiction, convinced that novels were overheating the minds of young people. In 1811 the Luddites in England broke the machines that were taking over their looms, sure that the machines meant the end of their craft.

Through the 1980s India’s bank unions fought the computerisation of banks with strikes and protests, fearing it would take jobs and break the way banking had always worked. Today you transfer money while standing in a queue for chai and nobody remembers that it was ever a fight. (Figure 2)

The pattern repeats so reliably you could set a clock by it. A new world arrives, people fear it, then it becomes so ordinary that nobody can remember being afraid. The fear never stopped the change. It only decided who adapted early and who adapted late. (Figure 3)

0404The fear has a second act

Berkeley Dietvorst, Joseph Simmons and Cade Massey (2015, Journal of Experimental Psychology: General) found that once people watched an algorithm make a mistake, they abandoned it faster than they abandoned a person who made the same mistake. They called it algorithm aversion.

So trust in the machine will wobble. There will be a viral wrong answer, a scandal, a season when everyone says AI cannot be trusted. Then, like every new world before it, it will settle into something ordinary and permanent. The businesses that win are not the ones who wait for the wobble to pass. They are the ones who are already inside the answer when it does.

0505What the new world cannot read

A company that has done good work for forty years keeps its reputation in places no machine has ever seen. It lives in the dealer who has stocked them since 1987, in the purchase manager who says “we only buy from them”, in the founder’s memory of why the big client stayed after a bad year.

The philosopher Michael Polanyi (1966, The Tacit Dimension) put it in six words: we know more than we can tell. A great deal of what makes an old business great was never written down, because nobody needed it written down. Everyone who mattered already knew.

In the old world that was an advantage. A newer competitor could build a better website, but it could not borrow forty years of people vouching for you. In the new world the most valuable thing you own is invisible to the voice your buyers now ask first. (Figure 4)

Look at where the answer engines take their information from. The same DareAISearch study reported that about 30% of AI citations came from publishers, about 25% from communities and user content and about 20% from video, with Reddit and YouTube at the top of the trusted domains list. Very little of it is a company’s own website. Most of it is other people talking, in public.

0606Our finding: the public uncle

That means the gap between a family business’s real reputation and its AI visibility is not a technology gap at all. It is the distance between how much people say about you privately and how much of it is said where the machine can hear.

0707How to adapt

Every business that survived a new world did the same thing: it took what was already true about it and carried it across.

Take what customers already say about you on a phone call or at a wedding and get it said in public. A Google review that says what you actually did, not just five stars. A short video of a real customer explaining why they have stayed for fifteen years. A post from someone in your team showing how the work is really done, the part an outsider never sees. The dealer who has stocked you for thirty years saying so on camera.

This is where a family business has an unfair advantage. A new brand has to buy attention because it has nobody to vouch for it. You have decades of people who already would. You only have to ask them to say it out loud, somewhere it lasts.

0808A philosophical aside

Every new world feels like the end of something and it is. Socrates was right that writing changed memory. The bank unions were right that computers changed banking. What each of them missed was the thing that survived the change, which was the reason the old world worked in the first place.

Reputation has always been something other people carried for you. The uncle never kept a database, he kept a feeling about who had treated his family well and passed it on at the right moment. Your name lived in him. The new world has not changed that. It has only changed the listener. The machine is doing exactly what the uncle did, repeating what people said about you. The only thing it asks is that they say it somewhere it can hear.

So the world as you knew it is over. The thing that built your business in it is not.

0909Where it lands

Your reputation was always built by other people talking. Make sure some of them are talking where the machines can hear.

1010Sources

1. Verizon. (2001). As landline phone celebrates 125th anniversary tomorrow, wireless service is reaching new heights of popularity. Press release. Source for the landline (91 years) and wireless (17 years) figures, US customers.
2. Reuters. (2023, 2 February). ChatGPT sets record for fastest growing user base, analyst note. UBS analysis of Similarweb data; source for ChatGPT (two months), TikTok (nine months) and Instagram (two and a half years).
3. Blair, A. (2010). Too Much to Know: Managing Scholarly Information before the Modern Age. Yale University Press. Source for Gessner, Bibliotheca universalis (1545).
4. DareAISearch. (2026). SearchScore.AI Discovery Index: The State of AI Discovery in India, Q1 FY2026 to 27. Reported in CXOToday, 10 August 2026, and CXO Digital Pulse.
5. Dietvorst, B. J., Simmons, J. P., and Massey, C. (2015). Algorithm aversion: People erroneously avoid algorithms after seeing them err. Journal of Experimental Psychology: General, 144(1), 114 to 126.
6. Logg, J. M., Minson, J. A., and Moore, D. A. (2019). Algorithm appreciation: People prefer algorithmic to human judgment. Organizational Behavior and Human Decision Processes, 151, 90 to 103.
7. Plato. Phaedrus, 274c to 277a (the story of Theuth and Thamus).
8. Polanyi, M. (1966). The Tacit Dimension. University of Chicago Press.
9. Simon, H. A. (1956). Rational choice and the structure of the environment. Psychological Review, 63(2), 129 to 138.

Figure 1 (figures/paper08-fig1-ai-visibility.png) plots the 19.3% and 80.7% split reported in the DareAISearch index. Figure 2 (figures/paper08-fig2-every-new-world.png) is a timeline of the fears named above: writing (around 370 BC), print (1545), novels (late 1700s), machine looms (1811), bank computerisation in India (1980s), AI answers (2026). It plots no data. Figure 3 (figures/paper08-fig3-faster-worlds.png) shows time to 100 million users or customers, bars to scale; the landline and mobile figures are US customers and the app figures are global monthly users, so the measures differ. Figure 4 (figures/paper08-fig4-what-the-machine-reads.png) is an original 1&O schematic and plots no data.