Hacker News
new
|
past
|
comments
|
ask
|
show
|
jobs
|
submit
login
nradov
31 days ago
|
parent
|
context
|
favorite
| on:
OpenAI’s head of ethics leaves less than a year af...
It's not technically possible to
prove
model alignment. At best you can establish a probability of alignment within certain constraints.
madrox
31 days ago
[–]
"Prove" is a shorthand because of course it's a stochastic process. My point is that I would expect ethics to be part of the training and eval criteria.
Guidelines
|
FAQ
|
Lists
|
API
|
Security
|
Legal
|
Apply to YC
|
Contact
Search: