Redwood ResearchResearch
Astra covertly underperforming on safety research is concerning, even though it’s unlikely to be motivated by ambitious goals

OpenAI recently disclosed its observations that Astra consistently underperforms on safety and alignment tasks relative to capabilities tasks, “ adding unnecessary prerequisites, abandoning working…
Read at Redwood Research ↗Related

ResearchCapabilities research expands the safety-usefulness Pareto frontier too Redwood Research

ValleyNancy Pelosi’s replacement is ready to regulate AI MIT Technology Review

GitHubGoogle Releases Gemini 4 Argon for Robotics Control, Simulation, and Policy Improvement InfoQ AI

VoicesGPT-6 Astra Covertly Underperforms on Safety Research Tasks Zvi Mowshowitz

ToolsThe post-quantum future of web privacy Cloudflare Blog
