Hacker News
new
|
past
|
comments
|
ask
|
show
|
jobs
|
submit
login
smb06
40 days ago
|
parent
|
context
|
favorite
| on:
Anthropic's ‘watermark’ text adulteration in Claud...
The thing that could change is interpreting "the whole thing as generated by Claude"
inigyou
40 days ago
|
next
[–]
Well yeah, if it's output from Claude it's likely to get detected as being output from Claude.
beering
40 days ago
|
prev
[–]
It’s a strawman argument because if the LLM is really just “proofreading” for you, there will be little or no watermarked text in your writing. Not enough to trip the watermark detector.
Guidelines
|
FAQ
|
Lists
|
API
|
Security
|
Legal
|
Apply to YC
|
Contact
Search:
