On May 1, 2026, IAC quietly posted a notice on Ask.com: "we have made the decision to discontinue our search business." Ask Jeeves — later rebranded Ask.com — was one of the internet's original search engines, a pioneer of the question-answering format before Google's algorithm-driven approach made it look quaint. Its closure landed with a particular irony: The Verge's Liz Lopatto had been writing, that same week, about the "Ask Jeeves-ification of online search" via AI chatbots. The future killed the past. Then the future arrived wearing the past's clothes.
The Shift That's Already Happening
Usage data tells a story that search industry insiders have been watching with increasing anxiety. Year-over-year, the proportion of informational queries being directed to AI chatbots rather than traditional search engines has grown substantially. ChatGPT, Claude, Gemini, and Perplexity are collectively absorbing queries that would previously have gone to Google — not all queries, not even most queries, but a meaningful and growing slice.
The queries migrating to AI first are predictable: "explain how X works," "help me understand Y," "what's the difference between A and B," "write me a Z." These are synthesis questions, not navigation questions. For synthesis, an AI that can reason over its training data and produce a coherent explanation is genuinely better than a list of links to pages that might contain the answer.
Google has noticed. Search volume growth has slowed in categories where AI chatbots are strongest. The company that defined search for a generation is now in the uncomfortable position of defending its core product against a technological shift it helped create.
What AI Chatbots Do Better
The advantages of AI search over traditional search for certain query types are real:
- Synthesized answers — Instead of reading five articles to understand a topic, you get one explanation that synthesizes across sources. For learning and understanding, this is faster.
- Conversational follow-ups — You can ask a clarifying question without starting over. Search requires reformulating the entire query; AI chatbots maintain context across turns.
- No ads in the answer — Traditional search results are heavily influenced by SEO and paid placement. AI answers (at least currently) are generated from training rather than purchased placement. The signal-to-noise ratio for informational queries is often better.
- Complex multi-part queries — "Compare these three approaches for my specific situation" is impossible to express as a search query but trivial to express in natural language to an AI.
What Traditional Search Still Wins At
The case for traditional search isn't dead, even as it weakens:
- Fresh news — AI chatbots have training cutoffs. For breaking news, real-time developments, or information from the last few hours, traditional search with news integration still wins. Perplexity has partially bridged this gap with live web search, but the freshness advantage of traditional search remains real for fast-moving stories.
- Local results — "Restaurants near me open now" is still better served by Google Maps and local search than by any current AI chatbot.
- Specific pages — When you know you want a specific website — a government form, a company's official documentation, a specific news article — search is faster than asking an AI to find it.
- Shopping — Price comparison, availability, seller reviews — these require real-time database access that most AI chatbots don't have.
- Images and visual search — Finding images, reverse image search, visual product discovery — still search territory.
The SEO Disruption No One Has Solved
Here is the uncomfortable question at the heart of the AI search transition: if AI systems answer questions by synthesizing information from websites, what happens to the websites that were the source?
The traditional web worked because search drove traffic, traffic drove advertising or e-commerce revenue, and that revenue funded the production of the content that made the web valuable. If AI answers questions without sending users to the source websites, the funding mechanism for content production breaks down.
This isn't theoretical. Publishers across news, how-to content, reference materials, and long-form journalism have reported declining organic search traffic as AI Overviews (Google's answer synthesis feature) and direct AI chatbot usage absorb queries. The clicks that previously reached their pages are increasingly being captured upstream by an AI summary. The content is still being consumed — it's training data and synthesis fodder — but the economic transaction that compensated the creator is not happening.
Publishers vs. AI: The Legal Battle
The response from major publishers has been litigious. The New York Times' lawsuit against OpenAI alleges that training large language models on copyrighted articles without compensation constitutes infringement. Similar suits have followed from other publishers, image libraries, and authors' organizations.
The robot.txt problem has become acute: publishers can add AI crawler blocks to their robots.txt files, but doing so cuts them off from the AI training pipelines that might amplify their content and from the search integrations (like Bing's AI, which uses web retrieval) that send traffic. It's a lose-lose choice: be scraped without compensation, or opt out and lose whatever distribution benefit AI search provides.
Some publishers have pursued content licensing deals instead of litigation. Associated Press, several major news organizations, and some academic publishers have struck deals with AI companies for access to their archives. The terms vary widely and are not uniformly disclosed. Whether these deals adequately compensate for the traffic and economic impact of AI-driven query absorption is contested.
Google's Response: Become the AI Layer
Google's strategic response to the AI search threat is the AI Overview, formerly called the Search Generative Experience (SGE). The premise: if users want AI-synthesized answers at the top of search results, Google will provide them — and retain the user relationship, the ad real estate, and the search habit.
The execution has been rocky. Early AI Overviews included factual errors that received widespread coverage — notably, recommendations to put glue on pizza and to eat rocks for their mineral content (apparently sourced from satirical Reddit comments the AI failed to identify as jokes). Google has improved the quality and deployed more conservative approaches in sensitive categories, but the launch established that AI-synthesized answers at scale have real reliability problems.
Google's deeper challenge is structural. The company's revenue depends on users clicking through to websites, where advertisers pay for placement. AI Overviews that answer questions without clicks threaten the click-through economy. Google is essentially being forced to cannibalize its own business model to remain competitive, which is one of the harder strategic positions in corporate history.
Perplexity: A Possible Middle Ground
Perplexity AI has positioned itself as a synthesis-with-citations model: give you an AI-generated answer while showing you the sources the answer drew from, with links. This approach attempts to retain the user experience benefits of AI synthesis (a direct answer) while preserving the attribution and traffic benefits that content creators depend on.
Whether it works economically is uncertain. If users read the Perplexity summary and click the sources, content creators receive traffic. If users read the summary and don't click — because the summary answered their question — the attribution is present but the economic transaction still doesn't happen. The citations are a step toward crediting sources; whether they translate to meaningful referral traffic is a different question.
Perplexity has also faced accusations of aggressively scraping content despite robots.txt restrictions, creating a tension between its creator-friendly positioning and its actual data practices. The category hasn't resolved the ethical and legal questions around content sourcing.
What the Economic Threat Means for the Web
Google's advertising revenue is the financial infrastructure of the web as we know it. Not Google's profits specifically, but the entire system: Google sells ads based on search traffic, that traffic funds content publishers through both direct Google ads and the behavioral data that underpins the broader programmatic advertising market. If Google's search volume and click-through rates decline materially, the ripple effect touches every publisher supported by programmatic advertising.
The scale of this is hard to overstate. Google's ad revenue in 2025 was over $200 billion. A significant decline in search-driven click traffic would be one of the largest economic disruptions in media history — larger than the newspaper advertising collapse that followed the rise of digital media, because the scope is global and the speed of change is faster.
What Developers and Site Owners Should Do
Practically, if you run a website, the AI search transition changes what "search optimization" means:
- Optimize for being cited by AI — AI systems cite authoritative, well-structured, factually accurate content. The traditional SEO tricks (keyword stuffing, backlink schemes) are less relevant; genuine expertise and clear communication are more relevant.
- Structured data matters more — Schema.org markup, clear headings, explicit author information, publication dates, and factual precision all help AI systems accurately represent your content.
- E-E-A-T signals — Google's "Experience, Expertise, Authoritativeness, Trustworthiness" framework becomes more important as AI systems try to evaluate source quality. Real credentials, real authors, real editorial standards are increasingly differentiating signals.
- Build direct audience relationships — Email newsletters, apps, communities — any channel that doesn't route through search becomes more valuable as search referral becomes less reliable.
Free tools that provide genuine value — like the JSON formatter, regex tester, or hash generator on online-clipboard.online — tend to hold up well in AI-influenced search because they solve specific problems that AI chatbots will mention and link to rather than fully replace.
The Obituary That's Premature
Search isn't dead. Google isn't collapsing next quarter. The transition from keyword search to AI-mediated query answering is real and accelerating, but it's happening at the pace of large-scale behavioral change — measured in years, not months. Billions of people have deeply ingrained search habits. Search remains indispensable for navigation, shopping, local discovery, and fresh news.
What is definitively over is the era of the "10 blue links" as the primary answer to informational queries. Ask.com's closure is a tidy symbol for that transition. The company that pioneered question-answering search — "Ask Jeeves" was literally a butler answering your questions — was killed by a generation of AI that actually answers questions. The vision was right. The technology just arrived twenty years late, and with a different company's name on it.
The web built on ad-funded content, powered by search traffic, is facing its most fundamental challenge since search itself. How it adapts will determine what the internet looks and feels like for the next decade.