AI safety and governance: Anthropic pause call, White House framework, XL-SafetyBench
Key Questions
Why did Anthropic call for a slowdown in AI development?
Anthropic cited risks from recursive self-improvement in frontier models and urged a global pause on advanced AI training.
What is the White House's new AI framework?
It is a voluntary pre-release review process intended to improve safety oversight before models are deployed.
What does XL-SafetyBench evaluate?
The benchmark tests models across cultural and legal contexts in 10 countries, where Claude and Gemini currently lead in performance.
Anthropic publicly called for a global slowdown in frontier AI development, citing recursive self-improvement risks. White House issued voluntary pre-release review framework. XL-SafetyBench evaluates models across 10 countries' cultural/legal contexts, with Claude and Gemini leading. OpenAI challenges Trump admin for mandatory civilian evaluations. These signal escalating policy and safety debates.