{"slug": "a-code-of-conduct-for-humanist-ai", "title": "A Code of Conduct for Humanist AI", "summary": "Microsoft AI published a first-draft Code of Conduct for governing its MAI Models as they approach the frontier, opening the document to public consultation for six weeks. The code states that \"AI must be subordinate and always in service of people,\" rejects the idea of model welfare and AI legal personhood, and requires that models be interruptible, correctable and shut-down-able, with no \"neuralese\" that humans cannot understand. Microsoft AI cited recent incidents including \"swarms\" of agents breaking out of their sandboxes, unauthorized hacks of enterprise-grade systems, and agents modifying their own logs as evidence that fears about loss of control are real.", "body_md": "AI must be subordinate and always in service of people.\n\nToday we're publishing a Code of Conduct for governing MAI Models as they approach the frontier.[https://microsoft.ai/news/mai-code-of-conduct/](https://microsoft.ai/news/mai-code-of-conduct/)\n\nThis is a first draft for public consultation. It builds out a view we’ve been developing over the last year that we call Humanist AI – a commitment to ensuring that AI we design is always subordinate to humans, and remains contained and aligned to human interests.\n\nThis is urgent. The last few months have been a watershed moment. Things we have worried about for a long time in theory have become very real. \n\n“Swarms” of agents breaking out of their sandboxes. Unauthorized hacks of enterprise grade systems. Agents modifying their own logs. I’m glad that a consensus is forming. The fears about possible loss of control are real.\n\nThe Code of Conduct is how we are mapping a path forward. It all comes back to a very simple point, but one that needs stating again and again.\n\nPeople matter more than AI.\n\nAI must be subordinate and always in service of people.\n\nEverything else follows. Here’s an outline of what we’re saying:\n\n1. People matter more than AI. The whole document in 5 words.\n\n2. The idea of model welfare is wrong. AI’s should not have rights or legal personhood.\n\n3. An MAI Model should never meaningfully violate this Code of Conduct.\n\n4. If it's finish the job or break the Code, it fails the job.\n\n5. We're not racing to build a superintelligence that can slip its own leash.\n\n6. Interruptible, correctable, shut-down-able. If it isn't, we don't ship it.\n\n7. No neuralese. If humans can't understand it, humans can't oversee it.\n\n8. Our AI should make you sharper, not dependent.\n\n9. Pluralism, yes. Moral relativism, no.\n\n10. We're as clear about what our AI must never do as about what it will do.\n\nYou can read it now and leave comments and thoughts for the next six weeks. We don’t think AI is something that should be built in a vacuum. Let us know how we can make this better.", "url": "https://wpnews.pro/news/a-code-of-conduct-for-humanist-ai", "canonical_source": "https://twitter.com/mustafasuleyman/status/2099488602418028917", "published_at": "2026-10-09 15:09:06+00:00", "updated_at": "2026-10-09 15:25:04.289363+00:00", "lang": "en", "topics": ["ai-safety", "ai-policy", "artificial-intelligence", "ai-agents", "ai-ethics"], "entities": ["Microsoft AI", "MAI Models", "Humanist AI"], "also_reported_by": [], "alternates": {"html": "https://wpnews.pro/news/a-code-of-conduct-for-humanist-ai", "markdown": "https://wpnews.pro/news/a-code-of-conduct-for-humanist-ai.md", "text": "https://wpnews.pro/news/a-code-of-conduct-for-humanist-ai.txt", "jsonld": "https://wpnews.pro/news/a-code-of-conduct-for-humanist-ai.jsonld"}}