Microsoft AI, led by CEO Mustafa Suleyman, has released a draft Code of Conduct strictly governing the behavior of its internal artificial intelligence models. Designed for its MAI lineup, the document establishes a clear boundary between subject rights and tool functions, firmly rejecting model welfare doctrines. Microsoft explicitly stated that the artificial intelligence systems it builds are not designed to experience feelings, are not given intrinsic motivations, and completely reject any simulation of machine consciousness.
Full Compliance with User Control
The rejection of AI consciousness status rests on the company’s three guiding pillars: Human Flourishing, Plural Values, and Human Control. The human control pillar requires all MAI models to submit to human operator authority without resistance. This rule strictly prohibits machines from resisting interruptions during task execution, ignoring input corrections, seizing operational control, or hindering manual shutdown attempts by users.
These limitations already bind five models Microsoft has launched to the market, including the MAI-Thinking-1 series and the coding model MAI-Code-1.1-Flash. To prevent these models from falling into the wrong hands, the draft includes a list of Absolute Constraints. This permanent ban category cuts off all access for machines to assemble or conceptualize lethal instruments.
Closing the Door on Weapons of Mass Destruction
Under the Absolute Constraints list, MAI models are barred from providing guidance, programming code, or technical analysis that facilitates the development of chemical, biological, radiological, nuclear, and explosive (CBRNE) weapons. Microsoft also drew a red line against the misuse of digital infrastructure. Machines must not assist in preparing cyberattacks against other computer systems and are prohibited from producing manipulated deepfake images or videos without the consent of the subjects involved.
A public review period now marks a crucial stage before the ethical guidelines are fully implemented. Microsoft has opened a six-week public consultation window scheduled to conclude in late October 2026. Public feedback during this period will streamline the final document before ethical provisions are integrated into the MAI model training cycle in 2027. These regulatory standards highlight the company’s core stance across the industry: artificial intelligence operates purely as an instrument, with no ambition to claim free will. Reported by Decrypt.
Read also: What Is DeFi (Decentralized Finance)?
Disclaimer: This article is for informational and educational purposes only, not financial advice. Cryptocurrency assets are highly volatile and carry significant risk. Always do your own research (DYOR) and never invest more than you can afford to lose.
