Leading AI labs signal that uncontrolled acceleration of automated AI development poses a real risk and call for coordinated international braking through technical and regulatory measures.
Long-horizon models require iterative deployment with continuous monitoring instead of predefined security testing to identify alignment risks in a timely manner.
Anthropic and OpenAI pursue opposing strategies for AI regulation at the state level: Anthropic supports progressively stricter standards, while OpenAI seeks uniform regulatory frameworks.
Anthropic will make hidden request throttling in Claude transparent going forward but retains content restrictions, partly due to conflicts with the US Department of Defense over national security.
