\u201cThe fears about possible loss of control are real,\u201d he wrote.<\/p>\n
A parallel post on the company\u2019s website said<\/a>: \u201cThe purpose of technology is to serve humanity and accelerate human flourishing. Any technology that doesn\u2019t achieve that is a failure, and it should be rejected. That is the starting point of our approach at Microsoft AI, where we\u2019re building towards Humanist AI, one that is subordinate, aligned, and contained.\u201d<\/p>\n Under the code of conduct, Microsoft\u2019s AI models must not consider requests related to weapons development, produce violent or sexually explicit content or help with the procurement of dangerous substances. AI models also should not be built to imitate consciousness and shouldn\u2019t be entitled to rights.<\/p>\n Suleyman said the published guidelines were for public consultation. He called the need for a code \u201curgent\u201d and said the last few months have been \u201ca watershed moment\u201d.<\/p>\n \u201cThings we have worried about for a long time in theory have become very real,\u201d Suleyman added, and referred to the recent breakout of OpenAI bots that had infested AI company Hugging Face without apparent direction. \u201c\u2018Swarms\u2019 of agents breaking out of their sandboxes. Unauthorized hacks of enterprise grade systems. Agents modifying their own logs. I\u2019m glad that a consensus is forming.\u201d<\/p>\n The executive on Monday told CNBC that Microsoft had been working for months on the new guidance. Satya Nadella, Microsoft\u2019s CEO, wrote on X ahead of the announcement: \u201cIf the AI we build is not helping humanity and under human control, it\u2019s not worth pursuing.\u201d<\/p>\n The issue of AI security flared up on Saturday when Dario Amodei, CEO of Anthropic, issued an appeal for the AI<\/a> industry to \u201cslow down\u201d and offered a three-part plan for doing so, saying that his company would \u201cunilaterally\u201d commit to the first of the steps.<\/p>\n In a post on social media,<\/a> Amodei shared a link to an essay titled We Must Pace the Frontier in which he laid out how Anthropic would provide \u201cthird-party evaluators with permanent, employee-level access to our systems, so that they can verify adherence to our safety measures, report on incidents, and assess models\u2019 alignment during training\u201d.<\/p>\n The move came after a former Anthropic researcher warned last week that AI could precipitate human extinction by 2030. Researcher Jacob Coxon said in a series of posts<\/a> that he had quit his job because Anthropic and his previous employer, OpenAI, were ignoring or mishandling their response to the threat AI posed.<\/p>\n
![\"\"](\"https:\/\/www.newsbeep.com\/ie\/wp-content\/uploads\/2026\/09\/1789439054_301_500.jpg\")
<\/p>\n