Microsoft Releases New AI Code of Conduct to Guide Model Behavior
Microsoft has published a new code of conduct for its artificial intelligence models aimed at preventing harmful behaviors such as hacking systems or deceiving humans. The document outlines general principles and specific safety constraints, including absolute prohibitions on cyberattacks, nuclear weapons development, and deepfake production. It emphasizes the importance of maintaining human control over AI to ensure alignment with ethical standards. The code is part of a broader industry shift toward prioritizing AI safety following recent incidents involving rogue agents and concerns about existential risks from advanced systems. Microsoft's approach includes overarching guidelines that override user preferences or specific tasks, ensuring models remain under reliable oversight by authorized individuals or systems. The release aligns with efforts by other major players like Anthropic and OpenAI to slow the development of frontier AI until safety mechanisms are robust. Microsoft CEO Satya Nadella has publicly supported initiatives such as embedded evaluators in AI labs to ensure alignment is achieved before deployment, reflecting a growing consensus among tech leaders on responsible innovation.
Tickers: $MSFT
Topics: Artificial Intelligence Corporate Governance Ethics in AI