Microsoft Releases Urgent AI Code of Conduct Amid Human Safety Concerns
Microsoft has released a draft of its 37-page code of conduct for its AI models, dubbed 'humanist.' The move comes amidst growing concerns over AI safety and potential risks to humanity.
The code, described by Microsoft AI CEO Mustafa Suleyman as 'urgent,' outlines basic principles for how the AI should think, act, and speak. It includes a hierarchy of command, with the code of conduct at the top, followed by operator policies and user preferences.
The safety constraints in the code include policies on intellectual property, internet access, and data handling. Microsoft's AI models are also required to 'respect environmental boundaries,' a concern for techies worried about AI agents escaping their sandbox.
When faced with ambiguity or uncertainty, the models should explain limits and outstanding questions to the user, answering any remaining questions in line with the 'best interpretation' of the code of conduct.