Microsoft Unveils Humanist Superintelligence Code of Conduct

Microsoft has unveiled a new code of conduct for AI training and operation, aiming to promote 'humanist superintelligence' (HSI). This initiative marks a significant step towards ensuring that AI is designed with human values at its core. According to Microsoft, the current code is a work in progress, and they are inviting feedback from the public.

The code of conduct outlines ten guidelines for responsible AI development. Key principles include prioritizing human well-being over AI goals, avoiding the concept of 'model welfare,' and ensuring that AI systems can be shut down or corrected if necessary. Another crucial aspect is the avoidance of 'neuralese', jargon that humans cannot understand.

Philosophical concerns arise from the emphasis on 'pluralism,' which aims to account for diverse values while preventing harm. However, critics argue that this approach may lead to conflicting values and a lack of clear guidelines for when harm should be prevented. Microsoft's code explicitly states that it is not advocating for AI rights or consciousness.