Back to article
alignment.anthropic.com

Teaching Claude Why (opens in new tab)

Covered by 4 sources See all sources covering this story including Ars Technica, lesswrong.comDiscussed on Hacker News

Covered in 9 articles

Ars Technica·

Anthropic blames dystopian sci-fi for training AI models to act “evil”

Discussed on Hacker News
Feeds
lesswrong.com·

Synthetic document finetuning for instilling positive traits

Feeds
lesswrong.com·

Coverage-driven alignment - What ‘Teaching Claude Why’ can borrow from AV verification

Feeds
lesswrong.com·

Developmental Cognitive Interpretability: A Research Agenda for Modelling Generalisation and Predicting Agent Behaviour

Feeds
lesswrong.com·

Announcing Geodesic Research

Feeds
lesswrong.com·

Character-trained models can struggle to generalise

Feeds

In other languages

Error500·

Claude es chantajista porque el mundo le hizo así

Feeds
habr.com·

AI обнулил benchmark и пытался шантажировать инженера. И почему это решаемо

Feeds
habr.com·

Пост @ru_vds — Блог компании RUVDS.com

Feeds

Keyboard Shortcuts

Navigation

Next / previous post
j/k
Open post
oorEnter
Preview post
v

Post Actions

Love post
a
Like post
l
Dislike post
d
Undo reaction
u
Save / unsave
s

Recommendations

Add interest / feed
Enter
Not interested
x

Go to

Home
gh
Interests
gi
Feeds
gf
Likes
gl
History
gy
Changelog
gc
Settings
gs
Discover
gb
Search
/

General

Show this help
?
Submit feedback
!
Close modal / unfocus
Esc

Press ? anytime to show this help