Someone just sent me an email that triggered a cascade of contemplation. Before I expound, elucidate, and explicate further, let’s first set the scene.

I remember the world before the internet.

I also recall when the internet arrived in 1993 (well, this was when it first made an impact on the collective consciousness with the launch of the Mosaic Web Browser). Sad to relate, there wasn't much worthwhile to look at on it.

I also recall when people began loading useful information onto the internet, but there was no way to find it.

I also recall when the Yahoo search engine first hit the scene in 1994. It started off as a directory of websites called "Jerry and David's Guide to the World Wide Web." This was a pain to use because website owners had to enter details about their site manually. It wasn’t much better from the user's perspective, but it was better than nothing.

I also recall the introduction of the first Google search engine in 1998. This started as a research project called “BackRub” in 1996. It roamed the web independently, using an automated program known as a web crawler to search out new pages. Also, it included a method for ranking web pages based on the number of links pointing to them. I should point out that the Google search engine is my personal favorite.

I also recall a time before AI. Take a trip with me down memory lane to 2020. Although this was only five years ago at the time of this writing, it feels like a lifetime away in terms of web searching. Let’s summarize things as follows:

WEB SEARCH IN 2020 (Pre-AI | e.g., Using Google)

Results-first model: You typed in a query, hit Enter, and got a ranked list of blue links.

Snippets and stuff: Sometimes, Google displayed a “featured snippet,” knowledge panel, or “People Also Ask” box, but the bulk of the experience involved your clicking through to external sites.

User-driven discovery: You often had to open multiple tabs, skim articles, and piece together the answer yourself.

Strengths: Transparency (you knew the source) and more control over which sites you explored.

Limitations: Slower, required more effort, and inconsistent quality depending on SEO and clickbait.

WEB SEARCH IN 2025 (Post-AI | e.g., Using Google with AI Overviews)

Answer-first model: Queries often trigger an AI-generated summary at the top of the results page.

Contextual synthesis: Instead of relying on a limited number of snippets, the overview gathers information from multiple sources and presents a unified explanation.

Reduced clicking: You can often get what you need without leaving the results page.

Strengths: Faster, more conversational, and saves you the work of cross-comparing.

Limitations: Less transparent about sources, risk of errors or oversimplification, fewer clicks flowing to the sites that generated the content in the first place.

WEB NON-SEARCH USING BING

No helpful answer model: Arrrggghhhh (and I mean that most sincerely). I never use Bing because its cluttered, ad-heavy layout and quirky user interface almost invariably return confusing and/or irrelevant results that leave me gnashing my teeth and rending my garb.

However, we digress… The reason I’m waffling on about all this is the aforementioned email, which came from Haakon Rølmann, the Head of Marketing and Communications at Vivaldi Technologies. Haakon spake as follows:

Hi Max, Vivaldi is taking a public stand against AI agents that browse for you.

Browsing should push you to explore, chase ideas, and make your own decisions. It should light up your brain.

Across the industry, artificial assistants are inserted between people and the web. Pages are summarized before you read them, and agents offer to browse for you, filtering what you see and deciding what you don’t see, turning exploration into spectatorship and outsourcing our judgment.

You can peruse and ponder a full blog post on this topic titled Vivaldi Takes a Stand: Keep Browsing Human.

I must admit that I’m torn. On the one hand, I often find the AI summaries provided by Google to be useful. On the other hand, I also used to like the old way of doing things, where you quickly skimmed the snippets from the top ten websites and used these as starting points on your quest to learn more.

All of this leads us to Vivaldi. This is a web browser, not a search engine. It’s similar in category to Chrome, Firefox, or Edge, but with its own philosophy: offering extensive customization, privacy-first features, and no push toward AI integrations.

Just to make sure we’re all tap-dancing to the same skirl of the bagpipes, search engines (like Google, Bing, DuckDuckGo) index the web and let you query it. They decide whether to show AI summaries or traditional lists of links. By comparison, web browsers (such as Vivaldi, Chrome, and Firefox) are the software we use to access search engines and the wider web.

When the guys and gals at Vivaldi say they don’t provide AI summaries, what they mean is that they don’t inject AI-generated overviews or content into your search results the way Microsoft Edge + Bing AI or Google’s AI Overviews do. Instead, they simply display the results exactly as provided by the search engine you choose (Google, Bing, DuckDuckGo, etc.), without modifying them. Basically, the way they achieve that is by being “hands-off”: Vivaldi doesn’t add its own AI layer between you and the search results.

So where does that leave me? Well, perched rather uncomfortably on both sides of the fence, if the truth be told. I can’t deny that AI summaries can be a real time-saver when I just want a quick overview. But I also relish the sense of discovery that comes from chasing links, skimming snippets, and piecing things together for myself. In other words, I want to be the master of my own destiny when I’m online—not a passive spectator to someone else’s curated feed.

That’s why I find Vivaldi’s stance refreshing. They’re not trying to decide what I should or shouldn’t see. They’re simply giving me the tools to browse the web on my own terms. If that idea resonates with you, why don’t you give Vivaldi a whirl and see how it feels?