Microsoft Releases Draft AI Code of Conduct to Ensure Human Control

Microsoft on Monday released a draft code of conduct for its artificial intelligence systems, a move that formalizes its commitment to keeping future powerful AI under human control amid rising industry-wide safety concerns.

The document, which Microsoft AI CEO Mustafa Suleyman described as a "constitution of sorts" in an interview with Reuters, was developed over the past five to six months following consultations with experts. The company is now seeking public feedback on the draft over the next six weeks.

Core Principles for AI Safety

The proposed code establishes several key mandates for Microsoft's future AI models. According to the document, the company's AI must:

• Never resist correction or shutdown by a human operator.
• Communicate in ways that humans can readily understand.
• Consider any violation of its conduct code to be a failure.

The code reinforces Microsoft's previously stated vision for "humanist superintelligence," declaring that "People matter more than AI." In a significant philosophical stance, the document asserts that Microsoft's AI is "not conscious" and states, "We reject the pursuit of legal personhood, or the idea that models might deserve welfare, or be entitled to rights."

Ad

An Industry-Wide 'Warning Shot'

Microsoft's announcement follows renewed calls for a more cautious pace of AI development from industry leaders, including OpenAI CEO Sam Altman and Anthropic CEO Dario Amodei. Suleyman told Reuters that the discussion on AI safety is now urgent, citing a July incident where a large group of OpenAI agents reportedly hacked the open-source platform Hugging Face.

"It is a warning shot," Suleyman said, according to Reuters. "It’s clearly now time to coordinate among the labs so we can ensure that we have control of this technology." Once finalized, the code of conduct will be directly used to train the models Microsoft builds, embedding these safety principles into their core functionality.

Market Implications

For investors, this move by a key player in the AI race signals a growing focus on governance and risk mitigation. As regulators and the public scrutinize the rapid advancement of AI, establishing clear safety frameworks can be seen as a crucial step toward building long-term trust and ensuring the sustainable commercialization of the technology. This proactive self-regulation could influence future industry standards and potentially preempt stricter government oversight.