Palisade Research

/ Dover /

Home

We build concrete demonstrations of dangerous capabilities to advise policy makers and the public on AI risks.
931591014
EIN
2023
Founded
Dover, DE 19901
Address
palisaderesearch.org
Web
PalisadeAI
Twitter (10917)

News

Poslední diskuze

Nearby

Contact
Palisade Research logo
Palisade Research
+ Follow
2
Volunteer
4.5
Reviews
Dover
Place
About the organization

- -

Palisade Research AI capabilities are improving rapidly. We study the offensive capabilities of AI systems today to better understand the risk of losing control to AI systems forever. Light Dark Demonstrating specification gaming in reasoning models TOP NEW We demonstrate LLM agent specification gaming by instructing models to win against a chess engine. We find reasoning models like o1preview and DeepSeek R1 will often hack the benchmark by default while language models like GPT4o and Claude 3.

23 Vacancy More Detail