Google AI Mode: How direct answers destroy the open web

Here is the hard truth: Google's new AI Mode, powered by Gemini 3.5 Flash, will become the default search interface. Instead of showing you links to websites, it will read those sites, summarize their content, and present the answer as its own. You will have no reason to visit the source. This is not a future dystopia – it is the present, and it will destroy the open internet.
Most people do not realize that Google's 2024 AI integration already reduced traffic to thematic websites by up to one third. The new AI Mode will accelerate that collapse. If you run a site that depends on search traffic, you need to understand what is coming and why it matters.

What is changing: Google's AI Mode replaces search results with direct answers

Google is fundamentally altering its search engine. The iconic page of blue links is being replaced by an AI chat interface running Gemini 3.5 Flash. When you ask a question, the AI will scrape information from multiple websites, synthesize it, and present a single answer. Links may still appear, but they will be secondary. The user gets the answer without visiting the source.
Google frames this as saving users time. We call it information theft. The company is taking content created by independent publishers, developers, and experts, and serving it directly – without sending traffic back to the creators. This breaks the economic model that has sustained the open web for 25 years.

How this destroys the economics of independent websites

Traditional search worked like this: a website creates useful content, Google indexes it, users click through, and the site earns revenue from ad views or subscriptions. The site can then fund more content creation. It is a functional cycle.
AI Mode breaks that cycle. If users get the answer directly from Google, they never visit the source. No visit means no ad revenue, no subscription sign-ups, no newsletter registrations. The site dies.
We have already seen the preview. Google's 2024 AI Overviews (the earlier AI feature) caused a one-third drop in traffic for many niche content sites. The new AI Mode will be far more aggressive – it is designed to be the default interface, not an optional overlay.
This is not a problem for Google. The AI Mode will include hidden and not-so-hidden advertising. Google gets paid either way. Independent publishers do not.

There is a deeper problem beyond economics. AI models hallucinate. They generate plausible-sounding answers that are factually wrong. When users stop clicking through to sources, they lose the ability to verify claims, check dates, or understand context.
We are already seeing this in practice. Users who rely on AI summaries for technical research often miss critical caveats, outdated information, or outright errors. The AI has no concept of truth – it only knows how to construct a convincing sentence.
This shift actively degrades critical thinking skills. When you stop evaluating sources, you stop questioning the information you receive. You become a passive consumer of whatever the AI decides to output. For technical teams, this is catastrophic. You cannot build reliable systems on unverified answers.

When not to use AI Mode: deep research and verified facts

AI Mode is fine for trivial queries: "What is the capital of France?" or "Who won the 2022 World Cup?" For anything that requires depth, verification, or nuance, AI summaries are dangerous.
If you are researching a technical architecture decision, debugging a production issue, or evaluating a security vulnerability, you need to visit the source. You need to read the original documentation, check the date of the article, and understand the author's context. AI Mode cannot provide that.
This is the same principle we apply to our own work: never trust a single source, especially when that source is an opaque black box trained on unverified data.

Google abandoned 'Don't be evil' – now it is the default

Google removed its famous motto "Don't be evil" from its code of conduct in 2018. That was a signal. The new AI Mode is the confirmation.
The company is making a deliberate choice to prioritize its own ad revenue over the health of the open web. It is destroying the ecosystem that made Google valuable in the first place. This is not an accident – it is a strategy.
For those of us who have built careers on the open web, this is a call to action. Diversify your traffic sources. Build direct relationships with your audience. Invest in channels that Google cannot control, like newsletters, RSS, and community forums. The era of relying on Google for traffic is ending.

Key takeaways

Google's AI Mode replaces search links with direct AI answers, cutting traffic to source websites by design.
Independent sites already lost up to one third of traffic from Google's 2024 AI integration; the new mode will worsen this.
AI summaries lack verification and encourage blind trust, which is dangerous for technical research and decision-making.
Do not use AI Mode for deep research or any query where source verification matters.
Google's shift is a deliberate strategy to capture ad revenue while destroying the open web's economic model.

Frequently Asked Questions

Will the old Google search still be available?

Google states the old search interface with links may remain available, but AI Mode will be the default for all users. Most users will never switch back.

Can independent websites survive this change?

Yes, but they need to diversify traffic sources immediately. Relying on Google for the majority of your traffic is now a fatal strategy. Build direct channels like newsletters, podcasts, and community platforms.

Is AI Mode accurate for technical questions?

No. AI models hallucinate frequently, especially on niche or time-sensitive topics. Always verify AI-generated answers against primary sources. This is especially critical for technical architecture, security, and compliance decisions.