Meta has announced a broad set of new measures to combat the growing sophistication of online scams across its platforms, including investments in advanced artificial intelligence systems, new user-facing warning tools on Facebook, WhatsApp, and Messenger, and expanded partnerships with law enforcement agencies and industry peers for offline enforcement action. The announcement reflects an acknowledgment that scammers are continuously evolving their methods in ways that traditional detection systems struggle to keep pace with, and that a multi-layered response combining technology, user education, and cross-industry collaboration is necessary to meaningfully reduce harm.
On the artificial intelligence front, Meta has built advanced systems capable of analysing multiple signals simultaneously, including text, images, and surrounding context, to detect a broader range of sophisticated scam patterns at scale. These systems are designed to identify celebrity impersonation, fake fan sentiment, misleading profile descriptions, and fraudulent associations with public figures or brands, as well as deceptive links and domain impersonation techniques where users are redirected to webpages that mimic legitimate ones. For users directly, Meta is testing new alerts on Facebook to flag suspicious friend requests where accounts show signs of unusual activity such as few mutual connections or a stated location inconsistent with prior behaviour. On WhatsApp, the platform will now display warnings when behavioural signals suggest a device-linking request may be suspicious, a measure targeting scams where criminals trick users into linking the scammer’s device to their account under false pretences such as fake talent competitions or misleading Quick Response code scans. Advanced scam detection on Messenger is also being expanded to more countries, with the system alerting users when new conversations contain patterns associated with common scams such as fraudulent job offers, and offering the option to share recent messages for an artificial intelligence-powered scam review before further engagement.
On the advertising side, Meta is expanding its advertiser verification programme with the goal of ensuring that verified advertisers account for 90 percent of its advertising revenue by the end of 2026, up from 70 percent currently, covering the highest-risk categories to limit misrepresentation of advertiser identity. The scale of the scam problem that Meta is contending with is reflected in the enforcement figures it disclosed: in the past year alone, the company removed over 159 million scam advertisements, 92 percent of which were taken down before any user reported them, and removed 10.9 million accounts on Facebook and Instagram linked to criminal scam centres. Meta described an observable industrialisation of scam operations globally, with criminal networks operating across borders and targeting users across social media, messaging platforms, dating applications, and cryptocurrency services simultaneously, reinforcing its position that countering this threat requires not only internal technological investment but sustained collaboration with external partners in law enforcement and across the technology industry.
Follow the SPIN IDG WhatsApp Channel for updates across the Smart Pakistan Insights Network covering all of Pakistan’s technology ecosystem.
