3 ms·
There could be a role for AI here. Anyone working on this?
by amelius 4y ago
There could be a role for AI here. Anyone working on this?
- ben_w 4y agoThat's probably worse, certainly in the short term. Long term, use many uncorrelated solutions: with humans "you can fool all of the people some of the time, and some of the people all of the time", but with a single AI (as with all software) once you know how to fool it, you can simultaneously fool every single person in the world who is using it.
- amelius 4y agoOk, but aren't we already using AI in e.g. spam filters? Also, there is some asymmetry in this problem. E.g. a little more false positives (a detection, but no actual threat) is not so bad, so we can err on that side.
- ben_w 4y agoAs I read it, this is the stuff which is already getting past such filters.
- doodlesdev 4y agoThere are already AI for preemptively blocking DNS records that look suspicious, a few of the DNS services I'm aware that OP could (and should) set up for his father are NextDNS [0] or AdGuard DNS [1][2]. Apart from dynamic detection using AI and heuristics both of these services also ship with DNS-level blacklists, you may also blacklist or whitelist specific websites if any lists generate issues on safe websites. [0]: https://nextdns.io/ https://nextdns.io/ [1]: https://adguard.com/ https://adguard.com/ [2]: https://adguard-dns.io/en/welcome.html https://adguard-dns.io/en/welcome.html