3 ms·
I don't know how Twitter plans to implement this, but I believe they could try to do so in good faith and using automation. 1. Target identifies disputed photo
by lovemenot 5y ago
I don't know how Twitter plans to implement this, but I believe they could try to do so in good faith and using automation.
1. Target identifies disputed photo of themself.
2. Twitter issues a dispute ID / QR
3. Target sends in a fresh selfie with the dispute ID printed or on a screen alongside.
4. Twitter performs face analysis of both images and OCR of ID.
5. If faces match and ID is validated, Twitter takes down the disputed image.
- golemotron 5y agoAnd it will be weaponized immediately to destroy citizen journalism.
- lovemenot 5y agoAre you saying that any means for an individual to have their own image removed would be weaponisation? If so, I disagree. If you are saying that third-parties will scale up access to Twitter's automation in order to take down virtually anything they wish, well I don't see how the proposed system would be so readily exploited. To issue a take down, you'd need to be in possession of the face.