Microsoft sets a new standard for security and human control for its Artificial Intelligence
In a strategic move to ensure technological security, Microsoft unveiled its new provisional code of conduct for Artificial Intelligence (AI) models this Monday. This document marks a firm commitment by the company to keep the development of these tools under strict human supervision, thus responding to growing global concerns regarding the autonomy of intelligent systems.
The framework, titled ‘Code of conduct for a humanistic AI’, establishes fundamental principles where the priority is the well-being of people. Microsoft emphasizes that any system that cannot be effectively supervised by human beings must be discarded, ensuring that models always operate in the service of their users and do not develop independent goals.
People matter more than AI. Any system that cannot be under human control must be rejected.
Microsoft
Key restrictions for ethical development
Within this new compendium of rules, the technology company has defined clear limits to prevent misuse. Among the most notable restrictions are:
- Absolute prohibition on collaborating in the manufacture of weapons or facilitating access to dangerous substances.
- Prevention of the generation of violent or sexually explicit content.
- Blocking the promotion of eating behaviors harmful to health.
- Preventing systems from developing their own goals or attempting to hide operational errors.
- Prevention of forms of communication incomprehensible to humans, ensuring transparency.
This code is currently in a six-week public consultation phase. A final, updated version is expected to be published by the end of the year, serving as the mandatory roadmap for models that the company develops from 2027 onwards.
A global debate context on AI
Microsoft’s initiative comes at a time of intense discussion regarding the risks of the technological race. Leading voices in the sector have recently warned about the danger of accelerating development toward autonomous systems without adequate safeguards. Given this landscape, the debate over whether AI progress should be slowed down has gained momentum.
For his part, US President Donald Trump has expressed his stance in favor of maintaining US technological leadership against China, distancing himself from calls suggesting a pause in innovation. With this new code, Microsoft seeks to balance technical advancement with an ethical structure that prioritizes user safety and operational responsibility.
