While many of the various Google AI ventures have been hit or miss, some have major marketing implications. The new features of their generative search applications, in particular, have some useful ideas for eCommerce. For example, the use of AI models for suggesting products before they physically exist is an interesting venture that the company is exploring.

What’s New in Google AI Applications?

For the uninitiated, we have a prior page about AI search engines and the possibilities they present. Notable features compared to the older version of Google include:

• The ability to generate digital images using prompts (similar to AI content tools based on Stable Diffusion).
• Extended search results with additional info for more relevant results.
• Incorporated product reviews in more depth.
• Better user-generated content incorporation.

The new system offers enhanced search features built on Google’s Shopping Graph which allows the AI to extract relevant facts that users might need about their products. This system makes crucial review keywords more important when marketing products.

Another similar feature is how it now categorizes reviews and opinions to find them easier. This can make information sharing and review-sourcing far simpler for searches. This is possible by letting large language models analyse the appropriate sources, so we’ll have to keep track of what sources become most important. Such a system could make Yelp and Reddit or consumer reviews a more powerful marketing tool (in theory).

The new updates will let AI describe images and process them into variants. It’s basically a version of automatic prompt collection and reiteration with stable diffusion.

One of the most interesting features (that is yet to be added) will be the generated product recommendations.

Product Search With an AI Powered Boost

The AI-based search engine may be a very powerful tool for future suggestions. Aside from AI’s capacity as an information contextualization tool, Google is positioning it as a product creation and concept A/B testing method. By using various forms of data collection and AI image processing, the search engine can suggest products that don’t yet exist but the user can have them made if they so choose.

Image-based processing allows the engine to pick up what data is most searched for and what people recommend and create new product concepts using it. This will probably have some growing pains but presents a lot of possibilities. The usage and customer interest data could provide a lot of consumer insight for marketers.

Aside from that, the idea generation possibility for many sites is doubled. Entirely new levels of marketing data could become available through these AI-generated product concepts

Google Search AI Accessibility

Right now, while SGE is not available in all regions, they are expanding it to 120 countries and additional languages. Currently, availability is limited but there are a few hacks for accessing it in Europe. If you want to know how to turn on Google AI search in Europe or other regions without access, here’s a four-step breakdown:

• You first need to create a private Google account so you can access certain features.
• Use a VPN to alter your IP address and host country.
• Now you need to access “https://labs.google.com/search” and switch on SGE/generative AI in the Search section. See the image below for more clarity:

• Now when you return to the Google Search page (google.com and many of its subdirectories), you should be able to see a Generative search option.

Of course, not everyone may want this feature since it requires tons of data collection. To turn off Google AI search, you can access “labs.google.com/search” and toggle the “on/off” switch, much like the steps shown above.

As Google’s AI systems improve, they are set to add more features and make it widely available. We’ll have to keep an eye on how the search engine will change as these features roll in.