One of the most impactful AI market events of the past six months was Google’s launch of AI Mode. For those unfamiliar, it brings Google search into the realm of AI assistants like ChatGPT, including natural language and follow-up questions that build on each other… like a human dialogue.
This has several implications for Google’s evolution as a company, and how it has to disrupt itself to stay competitive in the AI arms race. We’ll dive more into that broader angle in a bit – for now, things got a bit more interesting with AI mode in that Google announced it’s integrating the capability with images.
Specifically, Google is combining AI Mode with multimodal search. The latter is Google’s longstanding play around searching across media formats, including text, images, and video. With multimodal search, you can launch a search using an image, then use text to refine the search and zero in on what you want.
For example, using Google Lens you can use an image of a jacket that you saw someone wearing by snapping a picture of it. You can then launch a product search and drill down into specifics (think: “the same jacket in blue”). It’s all about increasing the surface area of search to as many inputs as possible.
Query Fan Out
Now, bringing AI Mode into the mix, you can do the same thing but with all of the natural language enhancements and multi-part questions that the feature is known for. So with the same jacket example, you can continue asking questions until you zero in on the right product, and where to buy it.
The non-AI-Mode comparison is what you’ve likely done historically – start over with new searches after each one. This method was sometimes successful in zeroing in on a given need. But AI Mode makes it more cohesive and progressive, remembering past questions as it goes… again, like a human dialogue.
Going under the hood, Google accomplishes this using a technique called “query fan-out.” This relies on object recognition that Google has developed through years of knowledge graph training and image search. This gives it the ability to understand images holistically, as well as their contents.
Google provides the example of a bookshelf. Using Google Lens, you can capture an image of a row of books then ask “If I enjoyed these, what are some similar books that are highly rated?” You could then follow up again to say something like “I’m looking for a quick read… what book should I buy next?”
They key word there is “buy.” As all the examples above suggest – books, jackets, etc. – use cases for visually-driven AI Mode will be commercial. And it could have local outcomes, such as finding out where to buy something nearby. That boosts the monetization potential, and therefore Google’s motivations.
Front Row Seat
Back to the broader story around AI mode, it alone isn’t very groundbreaking – and follows OpenAI’s December-released ChatGPT Search. But is notable in that it’s Google: It’s disrupting itself to stay competitive in AI. The conversational UX replaces traditional SERPs… where Google makes its money.
The thinking is that Google would rather disrupt itself than have someone else do it – a lesson it learned as it disrupted the heck out of media companies over the past two decades. It had a front-row seat to learn what not to do when being disrupted, which is to slip into denial and cling to legacy models.
AI Mode is the result of that thinking, and Google’s biggest step so far in disrupting itself. It likely has a plan up its sleeve for being able to mitigate the disruptive effect on its business, or to counterbalance losses in some way. For example, it could work towards trading quantity of paid search for quality.
Put another way, AI Mode – by virtue of a conversational UX – has the ability to discern deeper levels of user intent. Google can now process full sentences instead of traditional short search queries. With that level of intent inferences, Google could provision higher-value premium leads for advertisers.
But that’s all very speculative. And to give credit where due, much of this thinking around Google’s ability to provision high-value leads and offers flows from our go-to AI expert, Yext CDO Christian Ward. We’ll have to watch closely to see how Google navigates the perilous waters of disrupting oneself…
Header image credit: Jamie Street on Unsplash
