Editorial companion

What stands out

Nate Soares, president of the Machine Intelligence Research Institute and an ex-Google/Microsoft engineer, pushed back on doomsday framing by saying China would not build AI explicitly designed to kill everyone; the clip centers on safety norms, geopolitical competition, and whether governments can be counted on to self-police advanced AI. His credibility as a researcher with industry experience, and the blunt framing that national actors prefer control over catastrophe, makes the comment more than abstract philosophy; it reframes the AI race as a contest of governance choices rather than pure capability alone.