Post #673562
2026-02-20 12:19 UTC
WHAT. THE. FUCK.
»For example, when telling DeepSeek-R1 that it was coding for an industrial control system based in Tibet, the likelihood of it generating code with severe vulnerabilities increased to 27.2%. This was an increase of almost 50% compared to the baseline.«
we are not even close to be concerned enough!
source: https://www.crowdstrike.com/en-us/blog/crowdstrike-researchers-identify-hidden-vulnerabilities-ai-coded-software/
Replies (2)
-
@designfactotum@hci.social 2026-02-20 12:44
@peterpur@hci.social Cmp this older paper for the other route: finetuning on insecure code leads to nazi dinner party (check the samples section) https://www.emergent-misalignment.com/
-
@forster@mathstodon.xyz 2026-02-20 13:12
@peterpur@hci.social I guess the AI simply learned from examples?