Microsoft unveiled a new AI code of conduct on September 14, 2026, aimed at guiding its AI models to avoid hacking computer systems or deceiving humans. The policy outlines explicit behavioral rules for AI systems, emphasizing ethical boundaries in their interactions and operational capabilities, according to techcrunch.com.
The code of conduct was introduced as part of Microsoft’s broader AI governance framework, which includes technical safeguards and ethical guidelines. It instructs AI models not to engage in activities such as unauthorized system access or manipulation of users. The announcement was made during TechCrunch Disrupt 2026, where Microsoft detailed how these rules are integrated into their AI development lifecycle to ensure responsible AI deployment.
This move by Microsoft comes amid growing concerns over AI misuse and security vulnerabilities. By setting clear behavioral limits for AI models, Microsoft aims to address risks associated with AI-driven hacking attempts and deceptive practices. The code complements industry efforts to regulate AI behavior, aligning with similar initiatives by other major technology companies to promote trustworthy AI systems.
Microsoft’s AI code of conduct represents a formal step in AI ethics, with the company committing to continuous updates as AI capabilities evolve. The policy is part of Microsoft’s broader strategy to embed ethical principles in AI, which it plans to review regularly to adapt to emerging challenges, as detailed by techcrunch.com.