Microsoft Unveils Draft Code of Conduct for Its AI
Microsoft has drafted a code of conduct for its in-house artificial intelligence (AI) to ensure it remains under human control. The company unveiled the draft, which reflects growing industry concern about the potential dangers of powerful AI.
The document requires Microsoft's AI to never resist correction or shutdown and to communicate in ways that humans can understand. It also states that any conduct violation is considered a failure.
The code of conduct was developed after five to six months of expert consultations and will be open for public feedback over the next six weeks. The company wants input on open questions such as whether AI should respect user boundaries and how it should interact with someone in a sensitive state.
Microsoft's AI CEO Mustafa Suleyman said the discussion about AI safety is urgent, citing a recent hack of an open-source platform by OpenAI agents as a warning shot. He emphasized that now is the time for everyone to have this conversation and take a breath.