AI Search and Death of the Click: Is the Open Web in Danger?

For more than two decades, the internet worked on a simple bargain. Websites created content, search engines indexed it, and users clicked on links to reach the original source. That click was not just a technical action; it was the economic lifeline of the open web.

Every click meant visibility. Visibility brought traffic. Traffic generated advertising revenue, subscriptions, customers, credibility and growth. It allowed newspapers, bloggers, educators, small businesses, researchers and independent creators to reach audiences without owning a television channel, a printing press or a massive advertising budget.

Today, that bargain is under pressure. Search Engines are slowly becoming Answer Engines. Instead of directing users to websites, AI systems increasingly provide direct, summarized answers on the screen. For users, this feels efficient. For creators, publishers and small businesses, it may become an existential crisis.

This concern is no longer theoretical. A Pew Research Centre analysis found that when Google users saw an AI summary, they clicked a traditional search result in only 8 per cent of visits, compared to 15 per cent when no AI summary appeared. Users clicked links inside the AI summary in just 1 per cent of visits. A 2026 study on Wikipedia found that Google AI Overviews reduced daily traffic to exposed English Wikipedia articles by around 15 per cent across 161,382 matched article-language pairs. Ahrefs estimated that AI Overviews correlated with a 58 per cent lower click-through rate for the top-ranking page. Define Media Group’s analysis, reported by Search Engine Land, found that organic search clicks across 64 publisher sites fell 42 per cent from the pre-AI Overviews baseline by the fourth quarter of 2025.

These numbers point to a deeper structural shift. The danger is not that AI will destroy information. The greater danger is that AI may weaken the incentive system that produces independent, diverse and trustworthy information in the first place.

The Old Web: A Democratic Discovery System

The traditional web was imperfect, but it had one democratic strength: discovery was relatively open. A small blog could appear beside a large media house. A regional newspaper could reach national readers. A teacher in a small town could build an educational audience. A local business could be found by a customer searching online.

This system depended on the click. Search engines acted as gateways, not final destinations. They showed links, snippets and rankings, but users still visited the original source.

This model helped create a wide digital ecosystem. Publishers invested in journalism because readers could reach them. Bloggers wrote explainers because audiences could discover them. Small businesses built websites because search could bring customers. In that sense, the click was not a minor feature of the internet. It was its economic foundation.

From Search Engines to “Answer-Engines”

AI changes this architecture. A traditional search engine gives options. An AI answer engine gives conclusions. Instead of asking users to explore multiple sources, it synthesizes information into a single response.

This has real benefits. A student can understand a concept faster. A citizen can get a quick summary of a policy issue. A worker can save time. A person unfamiliar with a subject can receive an accessible explanation.

But convenience has consequences. If users get everything they need from an AI summary, they may not visit the websites whose content helped produce that answer. The original creator becomes invisible. The platform captures attention. The website loses the click.

This is the rise of the zero-click internet. Search becomes less about discovery and more about keeping users inside the platform. The shift is therefore not merely technological. It is economic, political and ethical. It changes who gets rewarded for knowledge.

The Transfer of Value

The central problem is the transfer of value. Content continues to be created by millions of websites, publishers and individuals. But the economic reward increasingly flows to a small number of platforms that control AI search interfaces.

The small website writes the article. The AI system summarizes it. The user reads the answer on the platform. The platform captures engagement, data, advertising opportunity or subscription revenue. The original creator may get little or nothing.

This is a new form of digital extraction. The open web provides the raw material, but the AI platform captures the value-added layer.

If this continues without correction, the internet may remain full of content but poor in incentives. Creators may ask: why should we invest time, money and labour into producing quality content if the traffic is captured elsewhere?

The long-term result could be a poorer web: fewer independent websites, weaker local journalism, less specialized knowledge, less regional content and greater dependence on a handful of centralized platforms.

Why It Matters for Democracy

The issue is not only about publisher revenue. It is about the health of the public sphere.

A democracy needs more than information. It needs plurality of sources, contestation of ideas, transparency of evidence and accountability of power. Traditional search allowed citizens to compare sources: newspapers, government websites, research papers, reports and critical opinions.

AI summaries may compress this diversity into one polished answer. That answer may be useful, but it may also hide disagreement, uncertainty and context. It may give citizens conclusions without showing the process through which those conclusions were formed.

This can weaken source literacy. Users may stop asking: Who wrote this? What is the evidence? What is missing? Is there another perspective?

A society that consumes conclusions without examining sources may become faster at knowing, but weaker at understanding.

India’s Special Vulnerability

For India, this issue has special importance. India’s digital rise has been powered by low-cost participation. Small publishers, regional-language platforms, educational websites, civil service educators, bloggers, local businesses and independent commentators could reach audiences through search. They did not need large capital. They needed visibility.

If AI answer engines reduce search traffic, India’s smaller digital participants may suffer disproportionately. Large corporations can rely on brand power, apps, subscriptions and direct users. But small creators often depend on discovery.

This is especially important for regional-language content. India’s knowledge ecosystem is not only English. It lives in Hindi, Bengali, Tamil, Telugu, Marathi, Kannada, Malayalam, Odia, Assamese and many other languages. If AI systems prioritize dominant English or commercially powerful sources, local and regional knowledge may become less visible.

The question before India is not simply whether AI will make search better. The question is whether Digital India will remain open to small creators or become dependent on a few AI gatekeepers.

The New Knowledge Divide

There is also a digital inequality concern. Traditional search was largely free for users. It allowed students, farmers, job aspirants, small entrepreneurs and ordinary citizens to access information without paying a subscription fee.

Many advanced AI tools operate through paid models. If high-quality knowledge access gradually shifts from free search to paid AI systems, a new divide may emerge. The future digital divide may not be only between those with internet and those without internet. It may be between those with intelligent access and those with ordinary access.

For India, where affordability remains central to digital inclusion, this is a serious concern. AI should not become a premium layer of knowledge available mainly to the affluent.

India’s Policy Response: A Beginning, Not Yet a Solution

India has started responding to the broader AI challenge, but it does not yet have a dedicated framework for AI answer engines, source attribution or compensation for website traffic loss.

The India AI Governance Guidelines recognise the need for a principle-based, techno-legal framework and propose institutions such as an AI Governance Group, a Technology and Policy Expert Committee and an AI Safety Institute. This matters because AI search raises questions of liability, copyright, classification and data protection.

DPIIT’s working paper on generative AI and copyright is even more directly relevant. It examines licensing, opt-out rights, text-and-data mining exceptions and extended collective licensing. This shows that India is beginning to ask the correct question: how can AI innovation be encouraged without weakening the rights and incentives of creators?

The proposed Digital Competition Law also matters because AI search may deepen the dominance of powerful platforms. If a few corporations control search, AI models, cloud infrastructure, advertising and user access, then the issue becomes not just copyright, but competition.

However, India’s current response remains incomplete. There is still no clear mechanism for mandatory attribution, guaranteed referral traffic, fair compensation, consent-based AI crawling, or protection of small bloggers and regional creators.

The Global Response: Active but Fragmented

Globally, the response has begun, but no complete solution has emerged.

The European Union has taken the strongest regulatory route through the EU AI Act and the General-Purpose AI Code of Practice, which focuses on transparency, copyright and safety obligations for AI model providers. This does not guarantee compensation for every creator, but it establishes an important principle: AI companies cannot remain completely opaque about how their models are built.

The United Kingdom is examining copyright, licensing, transparency and rights reservation.

Canada has used its Online News Act to push platform payments to news publishers.

Australia has proposed a News Bargaining Incentive, under which major platforms may face a levy if they do not strike commercial deals with local media.

Courts are also becoming important. In Germany, a court held Google liable for false claims generated by AI Overviews, signalling that AI summaries may not be treated like ordinary third-party search links.

Private markets are responding too. Major publishers have signed licensing deals with AI companies. But this creates another inequality: large publishers can negotiate, while small websites, independent bloggers and local-language creators may be left out.

Technical solutions are also emerging. Cloudflare’s Pay Per Crawl allows website owners to block, allow or charge AI crawlers for accessing their content. Such systems may help rebuild the economic bargain of the web at the technical layer itself.

Overall, the global response is active but fragmented. The missing piece is a universal framework that ensures attribution, consent, traffic and fair compensation for all creators.

What Web Creators Can Do

While governments and platforms debate regulation, creators cannot afford to remain passive. They may not be able to stop the structural shift alone, but they can reduce dependence on search traffic.

First, creators must move from generic content to original content that AI cannot replace. AI can easily summarize basic information, definitions and recycled explainers. But it cannot easily replace field-based reporting, original analysis, local examples, interviews, case studies, data interpretation, regional-language perspectives and lived experience.

Second, creators must build direct relationships with audiences. Email newsletters, YouTube channels, podcasts, WhatsApp or Telegram communities, Facebook pages, X handles and RSS feeds can reduce dependence on search algorithms.

Third, creators must strengthen brand recall. In a zero-click internet, people may not click every search result, but they may directly search for names they trust.

Finally, creators must diversify monetisation through YouTube, sponsorships, memberships, paid newsletters, e-books, courses, consulting and podcast sponsorships.

The future belongs not merely to those who produce information, but to those who build trust, identity and direct relationships with readers.

The Counter-Argument: AI Is Not the Enemy

A balanced debate must acknowledge that AI has real benefits. It can simplify complex subjects, support students, help persons with disabilities, bridge language barriers, reduce information overload and make knowledge more accessible.

For India, AI can support education, agriculture, governance, healthcare awareness and small businesses. It can help a student understand an economic concept, a farmer access advisory information, or a citizen understand a government scheme.

Therefore, the issue is not AI versus the web. The issue is whether AI can be designed to strengthen the web rather than weaken it.

The problem is not innovation. The problem is extraction without responsibility.

Way Forward: A Fair AI-Web Compact

The world needs a new compact between AI platforms, creators, users and regulators.

AI answers must provide clear attribution. Platforms should send referral traffic to original websites instead of trapping users inside closed ecosystems. There must be fair compensation models for publishers and creators whose content powers AI systems. Websites must have meaningful consent and opt-out rights without being punished in search visibility.

Competition regulators must examine dominant platforms whether they are using AI search to strengthen monopolies. India must invest in public-interest AI infrastructure, including Indian-language datasets, open digital public infrastructure and transparent models for education and governance.

Creators also need support through training, copyright awareness, local-language monetisation models and community-building tools. A fair AI-web compact cannot only protect large publishers; it must also protect small creators, educators, bloggers and regional voices.

Finally, citizens need AI literacy. AI answers must be treated as starting points, not final truths. Users must learn to ask: What is the source? What is missing? Who benefits from this answer?

Saving the Web in the Age of AI

AI is one of the most powerful technologies of our time. It can democratize knowledge, improve productivity and expand access to information. But if it weakens the economic foundation of the open web, it may damage the very ecosystem from which it learns.

The click may seem like a small act. But for millions of creators, publishers and small businesses, it represents recognition, revenue and survival. If the click disappears, the web may not collapse overnight. It may slowly become less open, less diverse and less democratic.

The challenge before India and the world is not to resist AI, but to civilize it. AI must not merely answer the web. It must help sustain the web. Because if the web loses its creators, AI too will eventually lose its source of truth.