More than $470 million was stolen in scams that started with a text message last year, according to the Federal Trade Commission. Google is far from the only company using AI to try to combat scammers and stop them from reaching people’s inboxes. Google’s Kleidermacher says that the company is seeing “really positive impact” from using its machine learning systems to detect potential scam messages in real time. As the protections continue to mature, he notes that the underlying system could eventually proliferate beyond just the Google Messages app into third-party communication platforms. The company also said on Tuesday that it is in the early phases of testing ways to incorporate scam detection for phone calls, but the capability has not been widely deployed.