Google search—as we know it—seems to be dying. In 2025, “classic searches” were down 33%. For many of us, AI is the new front door to the internet. We’re moving from the “Search Engine” era to the “Answer Engine” era. AI Chatbots like ChatGPT, Gemini, and Claude are now handling millions of queries that used to go straight to a classic search bar.
The Cannibalization of Traffic
What’s more: Google Search itself is cannibalizing its own results with AI summaries. By pinning them at the very top, other results are less visible. And that has caused many websites to lose a massive chunk of their traffic overnight. AI search is convenient—But what exactly does it change for us?
AI fundamentally changes what information we see when we search online. Remember the good old Google search? A list of links, ranked according to factors like keywords, popularity, and source. Now AI interprets information and writes up an answer, seemingly tailored to your needs.
Amplifying the Majority, Hiding the Minority
But that AI-generated summary often amplifies larger voices and hides smaller minority aspects. Because AI is trained on dominant sources, the system mirrors the majority view and sidelines perspectives that appear less frequently online.
AI search is also creating a new kind of “information inequality”. Depending on what language you use, your results might vary wildly. If you aren’t searching in a “major” language, you might be left with shallow or outdated info.
And remember: AI makes mistakes—and it lies convincingly. The summaries can present speculations or errors with unwarranted confidence.
So: Always double-check the results—and don’t be fooled by legit-looking links. They might be taken out of context. Or they might not even exist at all.
The “Wild West” of AI Marketing
AI is the new gatekeeper—for information and for products. While information providers seem to be scrambling to keep up, some businesses are already racing to secure their spot in AI answers. The industry is currently in the “Wild West” phase.
Reportedly, companies are trying to trick AI by using shady tactics to get their brands mentioned. This includes:
- Self-serving listicles: Ranking their own product at #1. These articles are designed to look like neutral reviews, even though they are effectively ads written to manipulate the AI’s training data.
- Hidden prompts: Marketers embed AI-targeted phrases, hoping the model will pick up the cues and repeat their preferred product messages.
So: Just because an AI recommends something, it doesn’t mean it’s the best product. Double-checking is highly recommended here, too.
The New Playbook for Visibility
Now, without any tricks: How can websites increase their visibility the legitimate way? Well, the old playbook is unfortunately burning up. Keywords and backlinks used to be effective strategies to direct people to your pages. But in this new world, the focus is shifting: It is now much more important to influence how AI interprets, summarizes, and recommends information.
Ok—what does that mean? In a nutshell: your brand needs to be talked about everywhere—on social media like Reddit, in niche blogs, and preferably in news articles. If AI finds enough “proof” that you are the expert, it will tell the users you’re the expert. If you aren’t being mentioned in the places AI crawls, you effectively don’t exist.
Experts argue this will disproportionally disadvantage smaller pages and enterprises, who don’t have years of online presence and can’t afford experts to optimize their online appearance EVERYWHERE.
The High Cost of Convenience
To sum it up: Users get fast, instant summaries. But unless the underlying AI systems become more transparent and less biased, the convenience comes at a high cost: Less diversity, less accuracy, and—as websites lose traffic and money—less information on the open web.
