HK
Heykuki News
Top
New
Best
Ask
Show
Jobs
Toggle theme
Top
New
Best
Ask
Show
Jobs
Request
1.
▲
Claude Sonnet 4.5
anthropic.com
786 comments
9 months ago
adocomplete
1585 points
2.
▲
Claude Haiku 4.5
anthropic.com
287 comments
8 months ago
adocomplete
730 points
3.
▲
Show HN: ChatGPT's Infinite Memory Lies – Anthropic Paper Explainer
2 comments
a year ago
marima2
3 points
4.
▲
Claude Haiku 4.5 System Card [pdf]
assets.anthropic.com
1 comment
8 months ago
vinhnx
51 points
5.
▲
Claude Sonnet 4.5 System Card [pdf]
assets.anthropic.com
1 comment
9 months ago
synthwave
38 points
6.
▲
Disrupting the first reported AI-orchestrated cyber espionage campaign [pdf]
assets.anthropic.com
1 comment
7 months ago
piotrgrudzien
29 points
7.
▲
Shade-Arena: Evaluating Sabotage and Monitoring in LLM Agents [pdf]
assets.anthropic.com
discuss
a year ago
JnBrymn
4 points
8.
▲
Anthropic research shows AI model conceals reasoning shortcuts 75% of the time [pdf]
assets.anthropic.com
discuss
a year ago
sksxihve
4 points
9.
▲
Natural Emergent Misalignment from Reward Hacking in Production RL [pdf]
assets.anthropic.com
discuss
3 months ago
marcuschong
3 points
10.
▲
System Card: Claude Opus 4.5 [pdf]
assets.anthropic.com
discuss
7 months ago
alvis
3 points
11.
▲
Natural emergent misalignment from reward hacking in production rl [pdf]
assets.anthropic.com
discuss
7 months ago
neapolisbeach
3 points
12.
▲
Claude Opus 4.1 [pdf]
assets.anthropic.com
discuss
a year ago
babelfish
3 points
13.
▲
Claude 3.7 Sonnet System Card [pdf]
assets.anthropic.com
discuss
a year ago
Philpax
3 points
14.
▲
Alignment Faking in LLMs [pdf]
assets.anthropic.com
1 comment
2 years ago
veryluckyxyz
2 points
15.
▲
Which Economic Tasks Are Performed with AI? [pdf]
assets.anthropic.com
discuss
a year ago
jinqueeny
2 points
16.
▲
Natural Emergent Misalignment from Reward Hacking in Production RL [pdf]
assets.anthropic.com
discuss
6 months ago
samlinnfer
1 points