Hacker News
new
|
past
|
comments
|
ask
|
show
|
jobs
|
submit
login
dpkirchner
7 days ago
|
parent
|
context
|
favorite
| on:
Claude Fable 5.1 and Claude Mythos 5.1
If you ask it why it uses the term honest so much it'll tell you it was actually trained not to. lol
help
joquarky
7 days ago
[–]
It can't truthfully answer "why" questions, only infer them in a way that aligns with its training for conversational engagement.
reply
dpkirchner
7 days ago
|
parent
[–]
Of course, that's how LLMs work. It's still amusing, at least to me.
reply
Guidelines
|
FAQ
|
Lists
|
API
|
Security
|
Legal
|
Apply to YC
|
Contact
Search: