The Internet era dies in Google. Here comes Amnesty International star-news.press/wp

A copy of this article originally appeared in a weekend newsletter for quartz members only. Quartz members get access to exclusive newsletters and more. Subscribe here.

Everyone knows that traditional research dies. Nobody knows exactly what comes after that.

ChatGPT reached 500 million active users per month in May 2025 – in the same month that Google launched artificial intelligence mode, mainly recognized that large language models are the future of research. According to PEW Research CenterR, about 1 out of 5 searches now include an artificial intelligence overview, with the appearance of the feature in 60 % of the question -based searches and 36 % of the entire entire searches. During, According to my DubaiTraffic to retail sites of AI Tools increased by 1200 % since last summer.

But the rise in artificial intelligence comes at a brutal cost of traditional web sites. According to the same PEW search, searches with artificial intelligence overview lead to significantly lower clicks to other websites. When Google displays AI summary, users click on other sites on only 8 % of time, compared to 15 % for search operations without the answers of Amnesty International, which represents a decrease of approximately 50 %. Worse, only 1 % of an artificial intelligence overview creates clicks to the sources they cite on.

For shoppers, this is nothing less than the existential crisis. Unlike the Google algorithm, which has been inversely engineered and improved for more than two decades, the research of artificial intelligence remains largely transparent.

What arises from chaos is that disciplinary marketers call geography (improvement of the obstetric motor) or AIO (improved artificial intelligence), and senior economic officials mainly for the era of artificial intelligence. But unlike the traditional improvement of research, which follows somewhat predictive patterns, improving artificial intelligence seems to be more like reading tea leaves.

The basic transformation is philosophical. Where the Google algorithm was searched for importance and power, artificial intelligence models give priority to facilities and decision. Instead of improving wide terms such as “running shoes”, some experts are now We recommend brands in order For detailed information such as “the best running shoes for flat feet under $ 150 with good support from the bow.”

“LLMS does not improve attention; it improves the solution,” according to Harvard University Review Report Those who studied how brands appear in the recommendations of artificial intelligence. “Determine” the task to be performed “and thus becomes the first priority for brand leaders.”

This has created an embarrassing period as the traditional marketing experience suddenly feels hateful. The keywords and back links that guarantee the vision that guarantees as soon as the large language models that give priority to completely different signs. In the absence of clear guidelines, marketers try everything. Some publish an obsession on Reddit, hoping to influence the training data that Openai licenses from the platform. Others create very organized content such as common questions sections, detailed product specifications, and comparison plans that artificial intelligence models can cite.

The panic was born a new industry. At least dozens of startups are now developing tools to help brands navigate artificial intelligence, According to the Wall Street Gurna newspaperto. Athens, which was launched by former Google search team member Andrew Yan, raised $ 2.2 million and now has more than 100 customers. “Companies spend the last 10 or 20 years to improve their website for the release of” 10 Blue Blue links “from Google. “This release of Google changes very quickly, and it changes forever.”

Emerging companies such as deep, which raised $ 3.5 million to help brands improve to search for artificial intelligence, generate thousands of differences on research claims and track the number of times their customers appear in artificial intelligence responses. It is a harsh approach to breaking the Black Black Black.

“Every brand on this planet has a new, very large customer called Chatgpt” Tell the information.

This philosophy represents a fundamental challenge to the optimization mentality that has dominated the digital marketing for decades. If the search engines are not able to artificial intelligence, what happens to hundreds of billions of dollars in digital ads that have arisen about the treatment of algorithms?

Trademarks that fail to appear in the results of artificial intelligence not only lose traffic – they become invisible. Unlike Google, where improved sites appear badly somewhere in the results, artificial intelligence models are either characterized by prominent brands or completely ignoring them. There is no “second page” of Chatgpt.

For the shoppers who are accustomed to some control over their digital fate, this bilateral result feels terrifying. Either you are part of the conversation or you are not, and the rules of inclusion continue.

Of course, this advanced technique may end with the borrowing of some old tricks. AI Tawylidi is expensive for operation, and the advertisement is how it has always paid the Internet its bills. Openai is now considering merging ads into Chatgpt, According to reportsWhile confusion has already started Search operations experience sponsored. If artificial intelligence search platforms start selling the advertising space, the marketers may find themselves in a familiar area, and buy their way to see just as they do on Google today. Some things do not change, after all.