Not much discussion on hn from the few threads that come up in the search. But I want to highlight rini17 which links to a post from GrapheneOS that in turn lin

Not much discussion on hn from the few threads that come up in the search. But I want to highlight rini17 which links to a post from GrapheneOS that in turn links to the Google blog from Oct 2024. My 2c is that this style of on device monitoring to warn users/children is an expected evolution of parental content monitoring tools for the llm era. Graphine is right. As open source it would be much better. Privacy and trust concerns would be easier to check and give the opportunity to make tools that empower parents and children to decide how they should moderate content on their devices. Perhaps some communities are very worried about nudity where others are more concerned about hate speech. Having the choice on what to detect and how, or if, to act would be helpful. https://news.ycombinator.com/item?id=43203303 https://news.ycombinator.com/item?id=43203303 https://security.googleblog.com/2024/10/5-new-protections-on-google-messages.html https://security.googleblog.com/2024/10/5-new-protections-on...