Sep 6, 2026
OpenAI’s chief scientist argues that alignment, monitoring and defensive systems must advance alongside AI capabilities.
Pachocki proposes safety bars for continued scaling and coordination to slow development when needed.
Explore on Humanity Unfolds
