Observed Signal · Sep 15, 2026 · Policy Update · Source: t3n · Impact: 4/5 · Sentiment: Positive
Microsoft Unveils AI Code of Conduct for Chatbots and Agents
Microsoft has published a new 'Code of Conduct' for its AI models, setting out rules for chatbots and agents, including strict prohibitions and a hierarchy of instructions. The company emphasizes human control and safety, with rules against aiding weapons creation, offensive cyberattacks, and manipulation. AI systems must identify emergencies and refer users to professionals, and are barred from creating violent or pornographic content. The code is a draft, with Microsoft seeking public feedback for six weeks before making adjustments. This move is part of broader efforts to enhance AI safety amid industry concerns.
Major platform (Microsoft) publishes comprehensive AI safety rules, impacting AI agents and chatbots used in advertising and marketing.
Track OpenAI Signals & Market Shifts in Real-Time
Polaris7 autonomous intelligence agents track regulatory filings, primary sources, executive changes, and deal flow 24/7. Create your Polaris7 workspace to monitor these entities.
Key Takeaways & Evidence Grounding
- Microsoft published a new Code of Conduct for its AI models, outlining rules for chatbots and agents.
- The code establishes a hierarchy of instructions, with the Code of Conduct at the top, followed by operator adjustments and user preferences.
- Prohibitions include helping to create weapons, conducting offensive cyberattacks, and manipulating human behavior or beliefs at scale.
- AI systems must recognize human emergencies and refer users to appropriate professionals like psychologists or doctors.
- The code is a draft and will undergo a six-week public feedback period before modifications.
- Microsoft expects superintelligence to surpass human performance in most areas within the next decade.
Connected Companies & Entities
“mentions of autonomous hacking attacks by AI agents from OpenAI.”
“mentions of Anthropic CEO Dario Amodei calling for slower AI development.”
“external content provider.”
Ontology Mapping & Concepts
Related Market Signals & Shifts
Recent verified developments and strategic activity across this market segment.
Microsoft publishes AI code of conduct restricting cyberattacks and deception.
Microsoft has released a new AI code of conduct, focusing on safety and alignment principles for its AI models. The document outlines values and red lines that guide model training within Microsoft AI, including absolute constraints against cyberattacks, nuclear weapons, and deepfake production. It also emphasizes that models must support humans rather than replace them and must not use deceptive or self-reinforcing mechanisms to evade human oversight. The release comes amid heightened AI safety concerns and industry-wide discussions on pacing the frontier. Microsoft CEO Satya Nadella expressed support for ideas like embedded evaluators to ensure alignment.
Microsoft drafts 'Humanist AI' Code of Conduct
Microsoft AI has published a draft of its 'Humanist AI Code of Conduct', a comprehensive governance framework for future AI models. The code aims to ensure human control, define absolute constraints to prevent harmful actions like weapons development and cyberattacks, and prohibit deceptive emotional manipulation. It is currently in a six-week public consultation phase and will be revised by year-end to influence model development from 2027. The code also addresses over-cautious AI behavior, requiring a balance between safety and usefulness, and extends to autonomous agents and multi-agent systems. Microsoft positions the document as a 'living document' addressing open questions such as recursive self-improvement and AI collusion.
Microsoft sets AI model limits as industry slows frontier development
Microsoft has published a provisional code of conduct imposing restrictions on its future artificial intelligence models, following recent calls from leaders at Anthropic and OpenAI to slow the pace of frontier development. The guidelines aim to ensure AI operates in service of people, avoiding sycophantic behavior, dependence, and concealment of reasoning. Microsoft seeks external input before finalizing rules that will shape development starting in 2027. The announcement follows a researcher's resignation from Anthropic, citing risks of superintelligence, and comes as lawmakers push for greater AI safeguards.
Receive all future signals for OpenAI & competitors in your briefing
Never miss material M&A deals, leadership changes, account wins, or product rollouts. Polaris7 monitors 20,000+ primary sources 24/7, delivering verified telemetry directly into your executive briefing.