Hacker News
new
|
past
|
comments
|
ask
|
show
|
jobs
|
submit
login
h6d_100c
63 days ago
|
parent
|
context
|
favorite
| on:
Cybersecurity researchers aren't happy about the g...
Or if GPU companies detected you were trying to train a model and injected intentional numerical errors.
gzalo
63 days ago
[–]
Nvidia already did something similar with Lite Hash Rate (LHR), limiting performance on purpose just when running mining apps...
h6d_100c
63 days ago
|
parent
[–]
Well they did tell everyone explicitly and sell it as different SKUs. There's no Fable (Full ML) edition, just silent prompt injection.
Guidelines
|
FAQ
|
Lists
|
API
|
Security
|
Legal
|
Apply to YC
|
Contact
Search: