Loading
Loading
  • Home

  • Productivity

  • App tips

App tips

6 min read

AI will change how you search—here's how

By Briana Brownell · March 13, 2024
A hero image with an icon representing AI search and SEO

Imagine it's February 2034. Taylor Swift's V-Eras Tour is completely in VR, grossing a trillion dollars. Denis Villeneuve is filming Arrival 2 on the moon. And OpenAI's ChatGPT, the tool that started the AI craze more than a decade earlier, has remade the internet as we know it. 

What will that mean for how you will find information online, including snagging your elusive Taylor Swift tickets?

In the decade-plus that I've been working in the data science and AI space, I've seen first-hand how technology tools have been adopted—and thrown aside as a relic of history. Here are some ways that I think generative AI tools will change how people interact with search engines.

The kinds of things you will search for will change

In 2002, researcher Andrei Broder categorized search queries into three types:

  • Informational, where a user is seeking specific information, like "taylor swift tour dates"

  • Navigational, where a user wants to find a specific site, like "taylor swift official website"

  • Transactional, where a user wants to make a specific purchase, like "taylor swift concert tickets for sale"

Another way to think of these is "know, go, do."

Estimates vary, but most research suggests that informational searches make up a majority of all searches. But with AI tools widely available—and getting very good at finding and summarizing information—the number of informational searches that people do will drop, while the number that robots do will skyrocket. Remember that the AI tools search and review multiple sites when pulling together a response. If anything, the number of sites they'll review before returning the information to the user will continue to grow.

Some informational searches will be completely handled by new AI tools, and in those cases, no human will actually read the original content. There's an opportunity for AI tools to improve the quality of some deeper informational searches, like product comparisons and information-gathering. But for them to do that, they need to gather that information from somewhere—maybe from your website or blog.

Search will become a dialogue

Right now, using AI tools to search is a lot like asking an assistant to do something for you. They take a broader context around what you want to accomplish and then convert that into search terms. They use those terms to search, and then extract the information they think you're looking for from the results. These tools can also summarize web page content before you visit, enabling a deeper and more efficient research process.

In this scenario, the AI tools are probably going to be the ones to create the search terms, rather than you. So in the next decade, our current search skills might become obsolete. Techniques like adding site:reddit.com or putting exact phrases in quotes will no longer be required. Instead, the AI tool will use its own techniques, and working with the AI tool will be the new skill set everyone will need to learn.

The Google Search home page and the ChatGPT home page
For now, ChatGPT looks eerily like Google Search.

This new skill set will grow in complexity because it will become iterative. The chatbot nature of AI tools allows you to broaden and narrow your search using a process more like a dialogue, instead of making a single search or a series of successive searches.

In between searches, you can add additional layers of complexity through constraints or adding context to improve the results. You can use the search results and refine and iterate by noting items that you felt were more on-target, and then ask the tools to provide more results like that one. This iterative nature means that you have more options at your disposal to find what you need. 

Multimedia search will become important

Our current searches are predominantly text-based, even though image searches have been possible for some time. Initially, image search functionality was limited to identifying similar images, helpful for recognizing a famous landmark on your vacation but not as much for figuring out what's wrong with your wilting houseplant based on a picture. 

But AI advancements have significantly broadened these capabilities, offering image description and annotation features. This means you can now search using images within your queries, making it possible to obtain precise information about the images you're curious about. Ask ChatGPT to describe a picture, and it will do so—and with a lot of detail.

ChatGPT describing a picture of a Pekingese dog uploaded by the user

Voice commands have also become a viable method for conducting searches or executing tasks, thanks to improved interpretation skills of AI tools. This isn't a new concept, with attempts by tech giants like Apple's Siri and Amazon's Alexa. But AI's current ability to understand extended monologues—and even ask for clarifications—is a significant leap forward. 

The integration of image and voice search will make it easier to access information that was previously challenging to find through traditional search methods. Video is probably next, and I expect there to be a continuous improvement in these features.

The places you will search will change

Generative AI in Google Search

The revolution in search isn't confined to the capabilities of AI tools like ChatGPT—the very platforms we use for searching are also changing. Social search is on the rise, with more individuals turning to TikTok and other platforms to find the information they need, rather than relying solely on the broader web.

AI is increasingly becoming an integral layer across all different platforms, overlaying social media sites or other repositories of content, allowing you to search specific places like Reddit or WolframAlpha. Although this was possible to do before, AI tools make it much easier, and will make it easier to surface trends or aggregate credible information—from anywhere.

Human curation might reemerge…

The internet is overflowing with content, making the search for specific topics (especially niche or less popular ones) a daunting task.

This challenge isn't new—it dates back decades. Initially, we turned to trusted human curators for guidance, with platforms like fark.com serving as go-to sources for discovering world events. The advent of personalized feeds marked a shift toward algorithmic curation—it seemed like it was signaling the end of human-led content discovery.

But the pendulum is shifting back. I think that the recent popularity of newsletters shows the power of human-curated content. Partially this is because the algorithms sometimes struggle to show you what you actually want to see—plus, the human aspect of the internet seems so rare nowadays. The growing importance of authenticity suggests we might start to appreciate human curation even more.

…But algorithms will still dominate

Despite the appeal of human-curated content, the sheer volume of online material makes algorithmic intervention necessary. Even if we want human-curated content, there's so much content out there that, in order to sift through all of it, any one person would need to spend a prohibitive amount of time and effort. So algorithms aren't going away completely any time soon.

Currently, the workings of these algorithmic systems are often opaque, but there's a growing call for transparency. Although companies may be reluctant to disclose the intricacies of their "secret sauce," the demand for fairness and understanding in AI is likely to set new standards. Expectations are shifting toward greater openness about how these tools function, including insights into why certain content is recommended and—maybe, just maybe—allowing users to influence these preferences.

The promise of personalization

Personalization is the promise tech companies have been making since before the dot-com bust. I see a lot of talk about AI tools trying to personalize user content, and in some ways, we've seen this, like with OpenAI's GPTs. There's talk of extending this further, too. ChatGPT is working on adding "memory," so it can remember things like your dietary requirements when it gives you suggestions of recipes or your interests when it suggests what to do on your vacation.

I've seen personalization constantly promised with new AI tools, and some tools, like Personal AI, are going all in on it. Will we finally deliver on this promise? If so, the implications for user experience could be huge.

The future of search

In the future, all search interactions might be mediated through an AI tool. There's some speculation that this will fundamentally change how we access everything on the internet. We might even all have AI agents that go out and gather information for us automatically or execute certain tasks for us without our intervention.

AI tools could completely replace search engines, and they could even completely replace web browsers. Everything would be mediated through a variety of AI tools, and websites as we know them may no longer exist. This is all speculative, but it's certainly possible. Regardless of how things change, search in 2034 might be completely unrecognizable from what it looks like today. But don't worry, you'll still be able to get your Taylor Swift tickets—even if your AI agent can't stand in the bathroom line for you.  

Related reading:

Get productivity tips delivered straight to your inbox

We’ll email you 1-3 times per week—and never share your information.

tags
mentioned apps

Related articles

Improve your productivity automatically. Use Zapier to get your apps working together.

Sign up
A Zap with the trigger 'When I get a new lead from Facebook,' and the action 'Notify my team in Slack'