Approov Takes on Rented Device Farms and State-Linked Attackers with Cloud-Verified Mobile API Defense
Approov 2027 (Version 3.7) detects leased iOS and Android device farms and virtual devices attackers exploit for fraud.
Press Release Disclaimer: This is a press release distributed through the XPR Media network. It has not been independently verified by our newsroom.
![]()
Fraud against mobile apps has stopped looking like bots. Attackers now rent real-looking phones by the hour, present device attestation that appears legitimate, and replay requests that were genuinely signed. Today Approov, a leader in mobile app and API security, announced Approov 2027 (Version 3.7), which detects rented device farms, ties every API request to the single device that produced it, and verifies both in the cloud.
“These are real banks of mobile phones that attackers are renting & manipulating to conduct scraping, fraud, and other API attacks,” said Ted Miracco, CEO of Approov. “They can circumvent Apple’s App Attest and Google’s Play Integrity checks because the devices themselves are genuine. The problem is that a legitimate device can still be used for malicious purposes. Approov goes beyond conventional device integrity checks by detecting the signals associated with these device farms and blocking their attacks at the API level.”
Stopping the Rented Device Farm
Cloud-hosted Android phones rent for as little as $0.10 to $0.50 an hour, can present legitimate device attestation, and are used to open and operate accounts for payment fraud, according to research published by Group-IB in March 2026. Device fingerprints and behavioral scoring were not built for an attacker who rents the real thing. Approov 2027 adds:
- Device-farm and virtual-device detection. Approov identifies and blocks leased device farms and virtualized devices that attackers rent to pass as real customers.
- Per-device request signing. Each device signs its API requests with its own key, so a signature captured from one device cannot be replayed from another. This goes a step beyond token-based attestation by tying every request to the device that produced it.
- Detection of new attack tooling. Added detections cover the latest hooking and instrumentation frameworks.
- Automated-abuse signals. New signals help security teams identify and block or rate-limit automated use of their APIs.
- Trusted time. A new SDK method gives the app a server-adjusted timestamp that does not depend on the device clock.
- Stricter hardware attestation policy. Customers can choose to reject any Android device whose hardware-backed key attestation is not valid and unrevoked, with the decision enforced in Approov’s cloud and not in the app.
Why the Trust Decision Belongs in the Cloud
Many attestation products make or sign the trust decision inside the app and rely on code obfuscation or white-box cryptography to hide the keys involved. Anything that ships in an app can be studied on a device the attacker controls, and AI tooling is making that work faster and cheaper. Approov takes a different approach. Its patented attestation is verified in the cloud, no long-lived secret is stored in the app, and API keys reach the app only at runtime after it has proven itself.
“We believe cloud-native verification is why Approov offers the strongest protection available for mobile APIs,” Miracco said. “No control completely eliminates fraud. What attestation does is move the real cost of an attack from software, which scales for free, to hardware, which does not. Renting a device farm was the attacker’s answer to that. Approov 2027 is our counterattack on these device farms.”
“Token-based attestation proves the app was genuine when the token was issued,” said Jae Hossell, CTO of Approov. “Per-device signing proves that each request came from that same device. A signature lifted from one phone is worthless on another.
“A device in an attacker’s hands is the one place you cannot keep a secret. That is why Approov makes the trust decision in the cloud, and why this release takes on the rented device farm.”
Tested Against a State-Linked Adversary
This year an Approov customer operating high-value mobile APIs faced a sustained campaign by a well-resourced adversary assessed as state-linked. The customer’s APIs remained protected throughout. What Approov’s engineers learned about that adversary’s tooling is built into this release for every customer. Approov is not identifying the customer, its sector or its location.
Looking Ahead: From Apps to AI Agents
Behavioral bot detection scores how a caller behaves. AI agents can imitate human behavior on demand, which weakens that signal. Attestation asks a different question.
“The question behind this release is what, exactly, is calling the API,” Miracco added. “Enterprises are now asking the same question about AI agents, and per-device, per-request proof is the foundation we are building on to answer it.”
Availability
The Approov 2027 security rules were delivered to customer accounts over the air on November 16 and 17, 2026, with no action needed from customers. The Version 3.7 SDK will be available on November 20, 2026, together with full release notes covering performance improvements, use of Cloudflare’s global network from an app’s first launch, and added support for GrapheneOS. For more information, visit approov.com.
About Approov
Approov is a comprehensive runtime security solution for mobile apps and their APIs, protecting against automated attacks, credential stuffing, and API abuse. By verifying that every request comes from a genuine, untampered app running on a safe device, Approov stops attackers, human or AI, in their tracks while ensuring a frictionless experience for legitimate users. Approov is based in Edinburgh, Scotland, and Palo Alto, California.
View source version on businesswire.com: https://www.businesswire.com/news/home/20261118631868/en/
Media gallery