Hacker News
new
top
best
ask
show
job
The Normalization of Deviance in AI
(
embracethered.com
)
13 points
by
gurjeet
7 hours ago
1 comment
Terr_
4 hours ago
I worry that over the last N months, we haven't really reduced the correctness/security risks at all, we've merely worked out more places we can plug in the risky thing.