Microsoft AI chief Mustafa Suleyman said AI models must be aligned with human interests and described OpenAI's new safety incident report as a rather serious indication of how powerful these systems are becoming. -- CNBC 18-SEP-2026 09:39 AM ET