On March 19, 2026, Meta said it is using stronger AI to make Facebook and Instagram safer. The company wants AI to find serious problems, especially scams, faster and with fewer mistakes. In early tests, Meta says the new system stopped about 5,000 scam attempts a day that older review teams had missed. It also lowered views of scam ads and other serious bad ads by 7%. Meta says this AI can work in languages used by about 98% of people online, so it may help users in many places. (about.fb.com)
One big problem is “celeb bait.” These are fake ads that use the face or name of a celebrity to look real. People click, trust the ad, and may lose money or give away personal information. Meta says its AI is getting better at finding accounts that pretend to be famous people. In tests, reports from users about the most copied celebrities fell by more than 80%. Meta also says it now protects the images of more than 500,000 celebrities and public figures, and it is using AI to catch “cloaking,” a trick that hides a scam website during ad checks. (about.fb.com)
Meta is also adding an AI support assistant on Facebook and Instagram. It can help users report scams, reset passwords, and change privacy settings, often in less than five seconds, according to the company. This matters because scams are still a huge problem: Meta says it removed more than 159 million scam ads in 2025, and 92% were removed before anyone reported them. The company is also expanding advertiser verification, with a goal that verified advertisers will bring in 90% of its ad revenue by the end of 2026. This story shows an interesting point: AI is not only for making new content. It is also becoming a tool for safety. (about.fb.com)










