News for hackers. Stuff that matters.

cnames

I’m in the U.S. and don’t want to rely on this stuff either, but I don’t know that I’d really trust any sufficiently complex software. Anything can be compromised. Even if you build from source, your compiler might be compromised, or you buy a cool new usb peripheral and plug it in- boom! Compromised. Bought a new device? Compromised already. That printer you bought years ago? Compromised and in your secure wireless network. Your sniffer and firewall? Compromised. Firefox? Compromised. Tor? Compromised. Wikileaks? Compromised. Your dishwasher? Compromised. You drive out in the woods without any devices and live in a tent. Hiker comes along and takes photos. You and tent are compromised. Walking out in the middle of nowhere naked hiding under a bush my ass.
Re:Agentic Misalignment: How LLMs could be insider threats on Saturday June 21, 2025 @02:52PM (#44339829)
0 days without a word accident
Re:Agentic Misalignment: How LLMs could be insider threats on Saturday June 21, 2025 @09:32AM (#44337466)
> Never use AI to do atomonous work

> having them do atomonous work is short sighted

I also think they shouldn’t be doing atomonous work. Maybe autonomous work, but never atomonous.