🇪🇸 Spanish, 25 October 2026, AI safety
Microsoft CEO Calls for Emergency Brake on AI Models
Satya Nadella proposes separating AI models from infrastructure to enable external controls and ensure safety.
The CEO of Microsoft, Satya Nadella, has warned about the risks of AI, stating that AI models need an emergency brake to stop them when their behavior deviates from the expected.
He argues that the current approach to AI, where models are treated as black boxes, is problematic as they acquire more autonomy and start to perform complex tasks.
Nadella proposes separating the AI model from the infrastructure that coordinates its functioning, allowing for external controls and measures to be put in place to ensure safety and accountability.