Inside Apple's 2025 App Store Security: Key Stats and Insights

Introduction

Apple has released its annual update on App Store safety, revealing impressive figures that underscore its commitment to maintaining a secure ecosystem. In 2025, the company processed over 9.1 million app submissions, using a blend of human review and advanced AI to detect and block malicious activity. While no system is perfect—some fraudulent apps slip through—the data shows Apple's multilayered defenses are evolving to match bad actors' tactics. Below are key questions and answers that break down the latest numbers and strategies.

How many app submissions did Apple review in 2025, and how many were rejected?

Apple's App Review team evaluated a total of 9.1 million submissions in 2025. Of those, over 2 million were rejected—including more than 1.2 million new apps and nearly 800,000 updates—for failing to meet the App Store's Review Guidelines. The company uses a combination of human reviewers and machine learning to maintain quality and safety. This effort ensures that only compliant, secure apps reach users, even though the sheer volume means some problematic submissions may still bypass initial checks.

What methods does Apple employ to detect fraudulent apps and developer accounts?

Apple relies on a multilayered defense system that blends human review with advanced machine learning. The AI is trained to identify complex malicious patterns, analyze app similarity, and flag suspicious changes in updates. Human reviewers then focus on the most critical cases, leveraging AI-generated insights. In 2025, Apple terminated 193,000 developer accounts over fraud concerns, demonstrating the effectiveness of these systems. Additionally, the company blocked over $2.2 billion in potentially fraudulent transactions, protecting both customers and legitimate developers.

How does Apple handle fraudulent app reviews and ratings?

Apple processed 1.3 billion reviews and ratings in 2025. Using a mix of AI and human moderators, the company identified and blocked nearly 195 million fraudulent reviews from ever appearing on the platform. The AI detects spam, offensive content, and inauthentic patterns at scale, while human reviewers handle edge cases. Apple states that its Trust and Safety teams integrate AI throughout the entire moderation process, enabling rapid discovery and response to new fraud vectors.

What role does AI play in Apple's Trust and Safety efforts?

AI is central to Apple's 2025 App Store security. The company explains that AI-powered dashboards and rapid data analysis tools help accelerate the detection of new fraud schemes. For example, AI is used to spot malicious app clusters, compare code patterns, and predict potentially harmful updates. This allows human reviewers to focus on high-impact decisions, speeding up the approval process for legitimate developers while improving customer experience. Apple emphasizes that AI is not a replacement for human judgment but a powerful assistant.

Why do some fraudulent apps still pass Apple's review despite these efforts?

Despite its sophisticated systems, Apple acknowledges that no review process is perfect. Bad actors continuously evolve their methods, using techniques like code obfuscation, delayed malicious behavior, or social engineering. The sheer volume of submissions—9.1 million per year—means a tiny fraction may slip through. Apple responds by continuously improving its machine learning models and updating review guidelines. When fraudulent apps are discovered, they are removed promptly, and the developer accounts are terminated, as seen in the 193,000 terminations in 2025.

What improvements has Apple made to speed up legitimate app approvals?

Apple's AI systems are designed to accelerate the review process for legitimate developers. By automating the detection of common guideline violations and flagging only suspicious updates for human review, the company reduces wait times. Apple states that this approach helps developers get their quality apps and updates to users faster. The combination of AI and human expertise ensures a high-quality, curated storefront while maintaining safety standards. In 2025, this balance allowed Apple to process billions of transactions and reviews without major bottlenecks.