Great piece overall! I’m hoping AI risk assessment and management processes can be improved.
Anthropic found that Claude 3 didn’t trigger AI Safety Level 3 for CBRN, but gave it a 30% chance of doing so in three months
30% chance of crossing the Yellow Line threshold (which requires building harder evals), not ASL-3 threshold
Thanks for the spot! Have fixed
Current theme: default
Less Wrong (text)
Less Wrong (link)
Arrow keys: Next/previous image
Escape or click: Hide zoomed image
Space bar: Reset image size & position
Scroll to zoom in/out
(When zoomed in, drag to pan; double-click to close)
Keys shown in yellow (e.g., ]) are accesskeys, and require a browser-specific modifier key (or keys).
]
Keys shown in grey (e.g., ?) do not require any modifier keys.
?
Esc
h
f
a
m
v
c
r
q
t
u
o
,
.
/
s
n
e
;
Enter
[
\
k
i
l
=
-
0
′
1
2
3
4
5
6
7
8
9
→
↓
←
↑
Space
x
z
`
g
Great piece overall! I’m hoping AI risk assessment and management processes can be improved.
30% chance of crossing the Yellow Line threshold (which requires building harder evals), not ASL-3 threshold
Thanks for the spot! Have fixed