Hacker News
new
|
past
|
comments
|
ask
|
show
|
jobs
|
submit
login
skeptic_ai
15 days ago
|
parent
|
context
|
favorite
| on:
OpenAI agents carried out an undisclosed attack on...
Can anyone explain why they can’t put a fake internet between agents and real internet. So if anyone reaches the fake internet already trips the safety flag.
alienbaby
15 days ago
[–]
They effectively try to do something just like this, but getting it setup close to perfect is very difficult.
jeremyjh
15 days ago
|
parent
|
next
[–]
It isn't difficult at all. It is difficult to do it without significant cost and inconveniences for the people running the training. That's it.
xpct
15 days ago
|
parent
|
prev
[–]
Is this related to why ChatGPT sometimes links to dead URLs?
Guidelines
|
FAQ
|
Lists
|
API
|
Security
|
Legal
|
Apply to YC
|
Contact
Search:
