Microsoft CEO Satya Nadella is pushing a clear message: however advanced AI becomes, humans must always be able to stop it. He wants every action by advanced AI systems logged, serious failures made public, and the “emergency brake” kept in human hands.
Nadella wrote on X that any pursuit of superintelligence must rest on the principle that AI which isn’t helping humanity and isn’t under human control is not worth pursuing. He also reportedly urged Microsoft employees in an internal memo to keep AI under human control, pace development deliberately and expand third-party testing.His stance comes amid a wider debate on slowing AI development, sparked by Anthropic CEO Dario Amodei and reportedly backed by leaders such as Sam Altman, Elon Musk and Demis Hassabis. The common thread is that as machines grow smarter, human oversight must grow stronger.
