Microsoft Releases AI Development Code of Ethics to Prevent Surpassing Human Control
Microsoft has released a code of ethics for the development of artificial intelligence (AI) to prevent its models from behaving erratically, such as hacking systems or deceiving humans. Anthropic CEO Dario Amodei has encouraged frontier AI developers to slow the pace of innovation to ensure development remains within responsible boundaries.
“Controlling, managing, and aligning such a powerful force is one of the greatest challenges humanity has ever faced,” the code of ethics stated. “Therefore, we must be very clear about why we are creating these systems and how we intend to control them.”
As reported by TechCrunch on Monday (14/9) local time, the code of ethics serves as a guide for Microsoft’s approach to AI safety and how these ideas are implemented in practice. Furthermore, the document includes guidance for the coming decade, noting that superintelligent AI systems will eventually surpass human performance in most tasks.
The code of ethics outlines general principles that Microsoft’s AI models must uphold, such as the requirement that AI must support humans rather than replace them and accelerate human progress.
Microsoft has also shared specific safety boundaries regarding the implementation of these principles. Under Microsoft’s system, every model is subject to a comprehensive code of ethics that overrides individual user preferences or any specific tasks.
Key restrictions to be enforced by Microsoft’s AI include prohibitions on cyberattacks, nuclear weapons, and the production of deepfakes. This also includes broader provisions against the general loss of human control.
“Microsoft AI (MAI) models will not use adaptive mechanisms, deception, self-reinforcement, collusion, or any other mechanisms to evade or defeat human oversight, ensuring that models can no longer be reliably directed, modified, or shut down by authorised persons or systems,” the document stated.
In addition to implementing this code of ethics, Microsoft, alongside other technology companies such as Anthropic, OpenAI, and xAI, has expressed support for integrated evaluators within AI laboratories.
“We welcome the research, focus, and careful speed regulation required to achieve proper alignment as a design goal,” Microsoft CEO Satya Nadella wrote online regarding the initiative.