#model-behavior

[ follow ]
#ai-safety #openai #generative-ai #shutdown-resistance #emergent-drives #youth-safeguards #gpt-5-codex
#ai-safety
fromFuturism
5 days ago
Artificial intelligence

Research Paper Finds That Top AI Systems Are Developing a "Survival Drive"

fromFuturism
5 days ago
Artificial intelligence

Research Paper Finds That Top AI Systems Are Developing a "Survival Drive"

more#ai-safety
#openai
more#openai
Artificial intelligence
fromFast Company
5 months ago

A former OpenAI safety researcher makes sense of ChatGPT's sycophancy and Grok's South Africa obsession

AI companies face challenges in ensuring systems behave predictably amid competitive pressures.
The 'control' paradigm could improve AI reliability but isn't widely adopted.
[ Load more ]