- Android trojans use TensorFlow AI to mimic human ad clicks for fraud.
- Fake apps on GetApps and other platforms spread malware with hidden browsers.
- Last count showed half a dozen apps, racking up more than 155,000 downloads altogether
Something strange is happening online – AI tools once meant for progress now help hackers fake user activity. Instead of real clicks, bots act like people browsing ads. These fakes slip past security systems designed to spot odd patterns. Money moves without anyone noticing the truth at first. Real companies pay for attention that never existed. The tricks evolve faster than the rules can catch up.
Clicks on advertisements bring income to both ad platforms and marketers. From the start of digital advertising, fraudsters have hunted methods to fake those clicks, using automation to inflate view counts and collect payments. Bots replaced real users, quietly boosting numbers behind the scenes. The lure of effortless profit drove constant upgrades in deception techniques.
Fake clicks always follow a pattern because machines run them, so ad systems started watching behavior instead. If clicking speeds feel off, too steady or oddly alike, those hits get tossed out. Some sites shuffle where ads show up each time, making bots miss their mark.
Fake apps to power the fraud
Today some fresh Android malware uses TensorFlow’s AI tools to study how people tap ads – then copy those moves closely. These fake clicks feel more natural because the system learns real user patterns over time. Instead of random taps, they act like actual users browsing apps. Machine learning helps them adjust timing, pressure, even swipe paths. The goal is slipping past detection by acting less robotic. Each interaction gets shaped by prior examples until it blends in. Not every attempt works perfectly, yet enough succeed to cause concern. Security teams now face smarter threats that evolve quietly. What once looked suspicious might today pass as normal activity.
Now running straight through sight alone, these fresh tricks skip old script paths entirely. Guided by patterns spotted in pixels, they lean hard into smart mimicry made possible by code that learns. Out in the wild, scammers tap a public toolkit – TensorFlow.js – to shape brain-like systems right inside web pages. This setup lets them host clever programs where people browse, or push them behind screens on distant machines via Node.js.
Criminals slipped malicious software onto Android phones by designing counterfeit applications. These bogus programs popped up not only on GetApps – Xiaomi’s legitimate app store – but showed up elsewhere too. Security analysts spotted them floating around independent web pages. Some spread through social networks. Others traveled via chat systems like Telegram.
A hidden browser runs inside certain apps, working silently in what’s known as phantom mode. Ads load without being seen, tucked away from real user view. This setup sits on a simulated display, out of sight but still active. Images captured from that display travel to TensorFlow. There, they get studied so ad locations can be spotted.
Start getting the TechRadar Pro email updates if you want sharp insights, real talk on tech trends, hands-on reviews, plus practical tips made for teams that aim to stay ahead. Success doesn’t shout – it shows up quietly in your inbox every morning.
Now the taps on screen buttons seem smoother, slipping past old-style security that watches how users act.
Folks mentioned the malware might broadcast the virtual browser display nonstop to hackers – so they’re able to poke around, swipe through pages, drop inputs whenever. Access stays wide open without pause.
Finding six apps so far adds up to over 155,000 downloads combined. While each operates separately, their total reach becomes clear when seen together. Numbers like these show how quickly usage can grow across different platforms.