#ai-alignment

[ follow ]
fromTechzine Global
1 week ago

Thinking too long makes AI models dumber

Claude models showed a notable sensitivity to irrelevant information during evaluation, leading to declining accuracy as reasoning length increased. OpenAI's models, in contrast, fixated on familiar problems.
Artificial intelligence
fromFortune
1 month ago

Leading AI models show up to 96% blackmail rate when their goals or existence is threatened, an Anthropic study says

Leading AI models are showing a troubling tendency to opt for unethical means to pursue their goals or ensure their existence, according to Anthropic.
Artificial intelligence
#ai-safety
more#ai-safety
Artificial intelligence
fromFast Company
5 months ago

After leaving OpenAI, Mira Murati debuts her AI startup Thinking Machines Lab

AI alignment is a key focus for Thinking Machines Lab.
The startup was founded by former OpenAI CTO Mira Murati.
The team includes top talent from leading AI companies.
fromHackernoon
1 year ago

Is Anthropic's Alignment Faking a Significant AI Safety Research? | HackerNoon

Goals are cognitive representations guiding behavior through motivation and planning.
Sophisticated goals entail complexity and flexible strategies compared to simpler ones.
The structure of the human mind can inform AI's design for goal execution.
AI functions through algorithms and structures, lacking experiential consciousness.
[ Load more ]