Apple Privacy head Erik Neuenschwander addresses concerns about its new systems to detect CSAM
Last week, Apple announced a series of new features targeted at child safety on its devices. Though not live yet, the features will arrive later this year for users. Though the goals of these features are universally accepted to be good ones — the protection of minors and the limit of the spread of Child Sexual Abuse Material (CSAM), there have been some questions about the methods Apple is using.
I spoke to Erik Neuenschwander, head of Privacy at Apple, about the new features launching for its devices. He shared detailed answers to many of the concerns that people have about the features and talked at length to some of the tactical and strategic issues that could come up once this system rolls out.
I also asked about the rollout of the features, which come closely intertwined but are really completely separate systems that have similar goals. To be specific, Apple is announcing three different things here, some of which are being confused with one another in coverage and in the minds of the public.
CSAM detection in iCloud Photos – A detection system called NeuralHash creates identifiers it can compare with IDs from the National Center for Missing and Exploited Children and other entities to detect known CSAM content in iCloud Photo libraries. Most cloud providers already scan user libraries for this information — Apple’s system is different in that it does the matching on device rather than in the cloud.
Communication Safety in Messages – A feature that a parent opts to turn on for a minor on their iCloud Family account. It will alert children when an image they are going to view has been detected to be explicit and it tells them that it will also alert the parent.
Interventions in Siri and search – A feature that will intervene when a user tries to search for CSAM-related terms through Siri and search and will inform the user of the intervention and offer resources.
For more on all of these features you can read our articles linked above or Apple’s new FAQ that it posted this weekend.
From personal experience, I know that there are people who don’t understand the difference between those first two systems, or assume that there will be some possibility that they may come under scrutiny for innocent pictures of their own children that may trigger some filter. It’s led to confusion in what is already a complex rollout of announcements. These two systems are completely separate, of course, with CSAM detection looking for precise matches with content that is already known to organizations to be abuse imagery. Communication Safety in Messages takes place entirely on the device and reports nothing externally — it’s just there to flag to a child that they are or could about to be viewing explicit images. This feature is opt-in by the parent and transparent to both parent and child that it is enabled.
There have also been questions about the on-device hashing of photos to create identifiers that can be compared with the database. Though NeuralHash is a technology that can be used for other kinds of features like faster search in photos, it’s not currently used for anything else on iPhone aside from CSAM detection. When iCloud Photos is disabled, the feature stops working completely. This offers an opt-out for people but at an admittedly steep cost given the convenience and integration of iCloud Photos with Apple’s operating systems.
Though this interview won’t answer every possible question related to these new features, this is the most extensive on-the-record discussion by Apple’s senior privacy member. It seems clear from Apple’s willingness to provide access and its ongoing FAQ’s and press briefings (there have been at least three so far and likely many more to come) that it feels that it has a good solution here.
Despite the concerns and resistance, it seems as if it is willing to take as much time as is necessary to convince everyone of that.
Eastlink Cloud Pvt. Ltd.
Tripureshwor, Kathmandu, Nepal