iPhone users facing the rising threat of snatch-and-grab street thefts are getting a powerful new layer of defense. Apple is actively testing a background security feature in the iOS 27.2 beta that instantly locks a device the moment it detects it has been stolen from the owner's hands.
Dubbed "Apple picking," the brazen act of snatching an unlocked smartphone has become a lucrative enterprise for criminals, particularly in cities like London where thieves reportedly reject Android devices in favor of high-value Apple hardware. To combat this, the newly released iOS 27.2 Beta 2 includes an internal mechanism called AutoLock. As discovered by 9to5Mac, the system is currently running silently in the background to gather telemetry data, meaning beta testers cannot manually enable or interact with the locking component just yet.
How the AutoLock System Detects Theft
To prevent thieves from exploiting an unlocked device, AutoLock relies on a combination of hardware sensors and connectivity metrics. According to the beta code, the system triggers based on the following signals:
- Sudden, sharp acceleration indicating the phone was snatched.
- Loss of connection to a paired Apple Watch.
- A paired Apple Watch remaining unreachable.
- Prolonged loss of network connectivity.
- The device remaining unlocked beyond a certain interval.
Because relying on a single metric could lead to frustrating false positives, Apple is building safeguards into the AutoLock framework. The system will evaluate biometric authentication history, current app activity, and a backoff mechanism designed to suppress repeated lock events if the user is simply moving quickly or experiencing network drops.
Closing the Unlocked Window
While Apple previously introduced Stolen Device Protection to prevent thieves from changing Apple ID passwords, that safeguard still left unlocked, active apps vulnerable if the phone was snatched mid-use. AutoLock bridges this critical gap by slamming the door shut before a thief can access banking apps or personal messages.
By running the feature silently in iOS 27.2 Beta 2, Apple is likely training its machine learning models to distinguish between a genuine theft and a user simply tossing their phone onto a couch. This background data collection is essential to ensure the final release doesn't disrupt daily use with errant lockdowns.
