Natural emergent misalignment from reward hacking in production rl [pdf]assets.anthropic.com3 points·by neapolisbeach·8 months ago·0 comments
Whisper leak: a side-channel attack on large language modelsarxiv.org3 points·by neapolisbeach·8 months ago·0 comments
Finding a command injection vulnerability in my smart thermostat (CVE-2023-4212)hmcguinn.com2 points·by neapolisbeach·3 years ago·0 comments
Vm2 discontinued due to unfixable security issuesgithub.com2 points·by neapolisbeach·3 years ago·0 comments
neapolisbeach·last year·discussNsjail, firecracker, gVisor, or v8 isolates are all good options with different tradeoffs
neapolisbeach·3 years ago·discussIf you're interested in time-lock puzzles, drand [1] achieves this but through randomness instead of computational effort.[1] https://drand.love/