Tweet
Nov 11, 2021,11:12am EST
Gordon Kelly
iPhone owners are currently facing multiple threats, but Apple’s new CSAM detection system has generated greater fears than even the biggest hacks. And it just got worse.
In a shocking new report (via BleepingComputer), a team of researchers at Imperial College London have found fundamental flaws in the technology behind Apple’s CSAM (Child Sexual Abuse Material) detection system. Apple intends to launch CSAM across all iPhones and iPads running iOS 15, but the report states that it is simple for images to both evade detection and “raise strong privacy concerns” for users.
CSAM operates by comparing image hashes (IDs) of pictures shared privately between iOS users to databases provided by child safety organizations. If matches are found then authorities and, where appropriate, parents are automatically notified. In theory, this makes the system rigorous and private. The problem is Imperial researchers found the whole system (and all systems of its type) can be bypassed simply by applying a hashtag filter to any image.
Gordon Kelly
iPhone owners are currently facing multiple threats, but Apple’s new CSAM detection system has generated greater fears than even the biggest hacks. And it just got worse.
In a shocking new report (via BleepingComputer), a team of researchers at Imperial College London have found fundamental flaws in the technology behind Apple’s CSAM (Child Sexual Abuse Material) detection system. Apple intends to launch CSAM across all iPhones and iPads running iOS 15, but the report states that it is simple for images to both evade detection and “raise strong privacy concerns” for users.
CSAM operates by comparing image hashes (IDs) of pictures shared privately between iOS users to databases provided by child safety organizations. If matches are found then authorities and, where appropriate, parents are automatically notified. In theory, this makes the system rigorous and private. The problem is Imperial researchers found the whole system (and all systems of its type) can be bypassed simply by applying a hashtag filter to any image.