4 ms·
At first I thought the bit about seeing user-reported content was a nothingburger, but this is far, far worse: Artificial intelligence initiates a second s
by square_usual 5y ago
At first I thought the bit about seeing user-reported content was a nothingburger, but this is far, far worse:
Artificial intelligence initiates a second set of queues – so-called proactive ones – by scanning unencrypted data that WhatsApp collects about its users and comparing it against suspicious account information and messaging patterns [...] as well as terms and images that have previously been deemed abusive. The unencrypted data available for scrutiny is extensive. It includes the names and profile images of a user’s WhatsApp groups as well as their phone number, profile photo, status message, phone battery level, language and time zone, unique mobile phone ID and IP address, wireless signal strength and phone operating system, as a list of their electronic devices, any related Facebook and Instagram accounts, the last time they used the app and any previous history of violations.
So the WhatsApp AI can arbitrarily decide to unencrypt your chats and send them for review. This makes the recent event where their CEO attacked Apple's CSAM detection algorithm almost funny in hindsight: takes one to know one, eh?
E: Upon looking into this further, the source from ProPublica makes it clear that the proactive queue cannot decrypt your messages. This was unclear in the linked article and only clarified separately in the ProPublica one. Despite that, I think the point still stands: WhatsApp gathers a ton of your personal data, and can reveal that to authorities on request.
- throwaway525142 5y ago> So the WhatsApp AI can arbitrarily decide to unencrypt your chats and send them for review. Where do you read the "decide to unencrypt your chats" part? I could only find that they have to decide based on the unencrypted parts of a WhatsApp profile plus other unencrypted data.
- square_usual 5y agoThe article clearly mentions that the AI initiates the proactive queue: Artificial intelligence initiates a second set of queues [...] by scanning unencrypted data. Initiates pretty clearly signals that no user reported the users in this queue. Also, the article mentions before this that the workers have to look at two queues of tickets which have unencrypted content for the moderators to judge. I agree that this is somewhat unclear, but I think it's certainly talking about two different things. E: the ProPublica article clarified that this is incorrect: https://www.propublica.org/article/how-facebook-undermines-privacy-protections-for-its-2-billion-whatsapp-users https://www.propublica.org/article/how-facebook-undermines-p...
- seanieb 5y agoNo they can’t arbitrarily decrypt messages. This process happens AFTER a user has complained and forwarded the unencrypted messages to WhatsApps abuse team. You’re first instinct is correct this is a BS nothing burger by journalist that should know better. It’s a stain on Propublica’s reputation.
- square_usual 5y agoIt's disappointing to see so many people here read maybe the first 10% of the article and decide it's nothing, when there's so much damning about WhatsApp's business model in here. WhatsApp revealed complete metdata that led to a whistleblower getting indicted, they harvest reams of personal data about you, and their AI can decide to unencrypt your information if it feels like it. Even the bit about moderating user-reported content isn't entirely "nothing" either, because WhatsApp's messaging clearly says nobody, including WhatsApp, can read your messages, which is incorrect. This is bad and not holding WhatsApp to account for this, when this very community spent weeks raging at Apple, as weird.
- seanieb 5y agoThey can’t decrypt your messages. You would first need to send them your unencrypted messages before they can read them… please correct your comment. There’s enough misinformation in the world.
- square_usual 5y agoI've responded to this elsewhere: https://news.ycombinator.com/item?id=28467601 https://news.ycombinator.com/item?id=28467601
- seanieb 5y agoIdeally you could edit ALL of your incorrect comments. And put the correction on top.
- ssss11 5y ago“Battery level” Ah yes, the ol’ suspicious battery level