Google has shared an overview of how it’s increasing its usage of AI-based systems to detect fraudulent business listings, improve Search, and even create podcasts about random artifacts.
Yeah, I’m not sure that those are going to be rising up the podcast charts anytime soon, but it’s another experiment that Google’s trying out.
First off, on business listings, Google has provided a new overview of how it’s implementing more AI-based systems to detect fraudulent entities within its systems.
As per Google:
“Our improved detection system blocked or removed more than 70 million policy-violating edits to places on Maps. We also removed or blocked more than 12 million fake Business Profiles.”
Google has also shared an overview of how its AI processes are getting better at uncovering misleading business listings:
“For example, we trained a new model with the help of Gemini that identifies potentially suspicious profile edits. A business that changes its name from “Zoe’s Coffee House” to “Zoe’s Cafe” isn’t suspicious – but a business that suddenly changes its category from “cafe” to “plumber” is. Gemini only needs a few examples to learn what’s suspicious and can apply this knowledge across many business categories and languages. This new model has already helped us block thousands of suspicious Business Profile edits this year.”
Seems like something you probably don’t need AI for, but at scale, I guess, that’s another analytical element.
Google’s also using AI to monitor business reviews, and is now alerting users to suspect activity:
As you can see in this example, Google’s now warning business profile visitors if it detects unusual patterns of review actions, in order to catch out businesses that have purchased fake reviews.
Which is a significant problem, and is part of the reason why people now turn to Reddit for more trustworthy, human reviews of businesses. Maybe, then, this could be a positive step to help bring more trust back into Google’s business links.
The new alerts are now being displayed on listings in the U.S., U.K, and India, and will be expanded globally next month.
On Search, Google says that it’s expanding access to its “AI Mode” to all Google Labs users, after strong feedback from initial testers.
Google says that users are entering longer search queries via AI Mode, and using the function to conduct more complex tasks, like comparing two products, or planning a trip.
And now, it’s also adding image-based understanding to the process.
“With AI Mode’s new multimodal understanding, you can snap a photo or upload an image, ask a question about it and get a rich, comprehensive response with links to dive deeper. This experience brings together powerful visual search capabilities in Lens with a custom version of Gemini, so you can easily ask complex questions about what you see.”
As you can see in this example, you can now take a photo, then ask context-based questions of the system, in order to get a better understanding of different elements. The system will understand the query, and be able to identify different elements of an image to provide more nuanced replies.
On a somewhat related front, Google’s also incorporating AI into its Arts and Culture tools, with new features that offer insight into classical artworks, and ways to experiment with creative tools.
And finally, there’s also the aforementioned podcasts tool.
“With our experimental Podcasts feature, Gemini AI generates insightful audio episodes about selected cultural artifacts. You can learn about how the mighty brown bear, technically a carnivore, actually gets about 90% of its diet from plants. Or, follow ancient Chinese pottery’s journey across Eurasia via audio. It’s a new way to transform your commute or relaxation time into cultural exploration.”
Seems like an interesting approach. Not interesting, as in “these will be great podcasts”, but interesting as in “what the heck are these podcasts even going to be?” But it’s another experiment that Google’s exploring.
Some interesting AI developments, as Google continues to go about its own process of developing AI tools to compete with the big players in the race.