Google Is Working on AI-Powered Fraud Detection for Chrome
Fraud is everywhere now . It’s never been more important to stay alert, whether you’re checking your email, browsing the web, or answering the phone. But it’s not all about you: Google can use artificial intelligence to prevent you from becoming a victim of fraud – at least when using Chrome .
As noted by researcher Leopeva64 , Google appears to be testing a new fraud detection feature in Chrome. Of course, it’s 2024, which means the feature uses AI. The feature name, “Client-Side Brand and Intent Detection for Fraud Detection,” doesn’t alert you to this fact, but you see it in the description: “Enables on-device LLM output on pages for the brand and page intent query. ” LLM stands for Large Language Model that powers many of the AI-powered programs and services you use, and this feature appears to use that AI model to find sites that appear to be fraudulent.
To further clarify the purpose of this feature, Leopeva described it to Gemini, which broke down the explanation of this feature and boiled it down to this: Based on bot analysis, the feature runs LLM on your device to detect two things: misrepresentation of brands, such as when scammers copy real ones brands to trick you into accessing their sites; and Suspicious Intent, which looks for key indicators of fraud and phishing attempts. You should always take AI-generated answers with a grain of salt , but this seems like a good interpretation in my opinion.
This option is available in the latest version of Chrome Canary , a version of Chrome that Google uses to test new flags (experimental features). It is not yet clear whether this works, but it can be enabled. First, launch Chrome Canary, then go to chrome://flags. From here, look for “brand intent and pages for client-side discovery” (without the quotes, of course), then click the “Default” drop-down menu and select “Enabled.” Click “Restart” and you’re done.
Chrome isn’t the only browser working on AI-powered fraud prevention. Last week, Leopeva64 discovered a similar feature in Microsoft Edge , a “malware blocker.” Unlike Google’s more cryptic description, this literally says: “Allow Microsoft to use AI to detect potential technical fraud.”