Malware developers are using app stores as distribution vectors to promote apps that secretly inject/replace/click ads via remotely simulating browser interactions.
Seen here in some research done by the security team at Baidu, these new methods of uncovering headless browser sessions enable publishers to protect themselves from being subjected to unwarranted payouts. Check out the keynote at Blackhat USA 2022 covering overlooked as giveaways to false click-throughs.
Using machine learning algorithms for analyzing large amounts of browser fingerprint data, these models can identify patterns and anomalies that are indicative of fraudulent activity with even greater accuracy as a highly effective method for detecting and preventing both known and unknown forms of ad fraud.