Recognizing an image before it spreads
In November 2017, this blog reported on an approach intended to help prevent intimate images from being shared without consent on Facebook and Instagram. The idea was to recognize a previously identified image through a digital fingerprint, rather than rely only on someone reporting each new upload.
The original headline was deliberately provocative. The practical issue deserves a clearer explanation: people facing intimate image abuse need trustworthy reporting and prevention tools, not a suggestion to post private images publicly.
What a digital fingerprint does
An image-matching system can compare a fingerprint, often called a hash, with other content to identify potential matches. Its effectiveness and privacy properties depend on how the system is designed. The original article’s broad claim that images were never stored should not be treated as a guarantee covering every program.
A current example
StopNCII explains its process: it creates a fingerprint of selected content on your device and sends the fingerprint, rather than the image or video itself, to the service. Participating platforms can use it to identify matching content and act under their policies.
The service doesn’t remove content from the entire internet. Check its eligibility requirements, participating platforms, and reporting instructions before using it.
Keep the historical claim in context
The original post did not establish when a particular 2017 pilot would reach the Netherlands or exactly what action would be taken against uploaders. Those unanswered questions shouldn’t be turned into promises. Use the platform’s current abuse-reporting tools and the official instructions of any prevention service you choose.
