Hacker News
new
past
comments
ask
show
jobs
points
by
erikus
2 hours ago
|
comments
by
edot
1 hours ago
|
[-]
And OpenAI didn't try to
silently
degrade performance of their top model if its (extremely sensitive) safety sensors went off ...
reply
by
estetlinus
1 hours ago
|
parent
|
[-]
Anthropic is the silver lining keeping p(doom) below 1.0
reply