A Massive Fraud Machine Meets a Massive Shield
Apple’s latest fraud prevention report underlines the scale of the battle raging behind every app download. In 2025, the company says it blocked more than USD 2.2 billion (approx. RM10.1 billion) in potentially fraudulent App Store transactions, bringing its six-year total to over USD 11.2 billion (approx. RM51.7 billion) in fraudulent transactions blocked. The numbers highlight both Apple’s defensive strength and the relentless pressure from bad actors targeting mobile users. On the customer side, Apple deactivated 40.4 million accounts for fraud and abuse, from bot-driven chart manipulation to orchestrated fake reviews. It also intercepted 5.4 million stolen credit cards and permanently banned nearly 2 million user accounts from making future purchases. These figures show that App Store fraud prevention is not a niche concern but a daily, industrial-scale operation that shapes what users see, trust, and pay for.
Inside Apple’s App Review Security: AI Plus Human Eyes
To keep malicious software out, Apple leans on a layered app review security model that combines machine learning and human judgment. In 2025, App Review teams evaluated more than 9.1 million submissions, ultimately rejecting over 2 million apps and updates that violated App Store rules. Problems ranged from privacy breaches and hidden, undocumented features to copycat designs and outright scam behavior. Nearly 59,000 apps were removed for “bait-and-switch” tactics, where an app initially appears legitimate—a simple calculator or puzzle game—then changes functionality after approval. Apple also blocked 443,000 submissions over privacy issues and terminated 193,000 developer accounts tied to fraud concerns, while rejecting about 138,000 new developer enrollments. Together, AI systems flag suspicious patterns at scale, and human reviewers make the final calls, illustrating how automated detection and manual scrutiny must work together to filter the torrent of incoming software.
Fake App Accounts, Ratings, and the Fight Over Visibility
Fraud on the App Store is not only about payments—it is also about visibility. Bad actors create fake app accounts and manipulate rankings to push deceptive titles in front of real users. Apple reports deactivating tens of millions of customer accounts for fraud and abuse, including those used to deploy bot networks that inflate downloads or flood listings with fake reviews. Out of 1.3 billion ratings and reviews processed in 2025, nearly 195 million were filtered out as fraudulent. On top of that, Apple blocked nearly 7,800 deceptive apps from appearing in search results and stopped another 11,500 from artificially climbing the charts. By trying to neutralize these tactics, Apple aims to protect legitimate developers from being buried under manipulated results—and to ensure that users are not steered toward risky software by numbers that look authentic but are generated by coordinated fraud campaigns.
Where Scammers Still Sneak Through: Trials, Subscriptions, and Side Channels
Despite billions in fraudulent transactions blocked and aggressive policing of fake app accounts, scams still slip through Apple’s defenses. A growing problem involves apps that technically comply with guidelines but are designed to confuse. Developers hide recurring charges behind obscure trial terms or design interfaces that make it easy to accept expensive subscriptions by mistake. Bait-and-switch behavior is also rising, with tens of thousands of apps changing functionality after review, forcing Apple to remove them only once damage is underway. Outside the official storefront, Apple says it detected and blocked thousands of illegitimate apps distributed via pirate marketplaces, along with millions of attempts to install unapproved software. These gaps reveal the cat-and-mouse nature of platform security: even with advanced AI, human reviewers, and huge fraud prevention budgets, scammers continually adapt, leaving users to play a critical last line of defense by reading terms and reporting suspicious apps.
