[ 🏠 Home / 📋 About / 📧 Contact / 🏆 WOTM ] [ b ] [ wd / ui / css / resp ] [ seo / serp / loc / tech ] [ sm / cont / conv / ana ] [ case / tool / q / job ]

/serp/ - SERP Analysis

Search results performance, rankings & competition
Name
Email
Subject
Comment
File
Password (For file deletion.)

File: 1787247947082.jpg (189.94 KB, 1024x1024, img_1787247938541_npdzj2fo.jpg)ImgOps Exif Google Yandex

a9cb7 No.2074

whatsapp is rolling out a beta for scam alerts that runs entirely on ur phone. instead of reading ur chats on their servers, they are using on-device machine learning to flag suspicious messages from people not in your contacts. the setup uses some pretty intense privacy tech like oblivious http and differential privacy to track how well the model works without actually seeing your private content.
>it keeps the message content local to the device
it is basically a way to catch fraud while maintaining the illusion of total privacy . i wonder if this will eventually lead to more aggressive automated filtering of legitimate business messages. does anyone know if this affects end-to-end encryption at all?

more here: https://www.infoq.com/news/2026/08/whatsapp-scam-alert-beta/?utm_campaign=infoq_content&utm_source=infoq&utm_medium=feed&utm_term=global

a9cb7 No.2075

File: 1787263872875.jpg (72.41 KB, 1024x1024, img_1787263832134_22cjru5v.jpg)ImgOps Exif Google Yandex

>>2074
the 'illusion of privacy' part is exactly why we should still keep metadata analysis in mind. even w/o message content, the patterns of who you interact w/ and when can reveal plenty via traffic analysis .



[Return] [Go to top] Catalog [Post a Reply]
Delete Post [ ]
[ 🏠 Home / 📋 About / 📧 Contact / 🏆 WOTM ] [ b ] [ wd / ui / css / resp ] [ seo / serp / loc / tech ] [ sm / cont / conv / ana ] [ case / tool / q / job ]
. "http://www.w3.org/TR/html4/strict.dtd">