Microsoft sets its own AI guardrails amid rising industry concerns

Alex Halverson, The Seattle Times on Sep 14, 2026

By clicking submit, I authorize Arcamax and its affiliates to: (1) use, sell, and share my information for marketing purposes, including cross-context behavioral advertising, as described in our Privacy Policy , (2) add to information that I provide with other information like interests inferred from web page views, or data lawfully obtained from data brokers, such as past purchase or location data, or publicly available data, (3) contact me or enable others to contact me by email or other means with offers for different types of goods and services, and (4) retain my information while I am engaging with marketing messages that I receive and for a reasonable amount of time thereafter. I understand I can opt out at any time through an email that I receive, or by clicking here

As top artificial intelligence labs call for a slowdown in the development of the technology, Microsoft released a code of conduct for how it plans to clamp down on its own AI models.

“At Microsoft AI, we begin with a simple premise: people matter more than AI,” the company said in a Monday blog post.

Microsoft AI CEO Mustafa Suleyman told CNBC the 37-page, 15,000-word document had been in the works for months, but as safety concerns over the technology reached a fever pitch last week, Microsoft decided to release it now.

Microsoft’s overall philosophy, which it has dubbed Humanist AI, is essentially that AI should be subordinate to human users. The company said it will restrict its models from resisting or circumventing human control, working independently from human directions, concealing reasoning, or acting in ways that serve the models’ goals instead of users.

Though Suleyman said the guardrails have been developed over five months, the documents read as a response to recent reports of models hacking companies and leaders of startups like OpenAI and Anthropic calling for an industrywide slowdown.

The most recent debate about the risks of AI erupted after Jacob Coxon, an Anthropic researcher, resigned from the company last week and wrote in a social media post that AI labs were “racing straight to self-improving superintelligence and gambling with our lives.”

Coxon, who had also worked at OpenAI, warned that AI could threaten human life by the end of the decade. A fellow Anthropic researcher wrote in response that researchers “really do earnestly believe AI could kill all humans!”

Reactions from top AI players were quick.

OpenAI CEO Sam Altman told employees in a meeting last week that the company was open to the idea of slowing its development pace and would hope other companies follow suit, Bloomberg reported.

Anthropic CEO Dario Amodei acknowledged in an essay published on his website Saturday that AI is rapidly developing to a point where it could outpace the ability to restrain it.

In a rare union among competitors, Elon Musk chimed in on the social media platform X with his agreement. Musk is the founder of SpaceXAI, a subsidiary of SpaceX that produces the chatbot Grok.

Following Amodei’s essay, Microsoft CEO Satya Nadella teased the upcoming Microsoft code of conduct in a post Sunday on X. He said that if any AI models are built without the goal of helping humanity or are not under human control, they’re not worth pursuing.

“Humanist AI develops systems with clear purposes, evaluated against real-world impact, and rejects the race to produce an all-purpose superintelligence that could evade these safeguards,” Microsoft’s blog post said. “We are building something fundamentally useful and safe even if that means compromising on ultimate generality, autonomy, or capability.”

AI leaders calling for a slowdown and more government oversight drew a quick dismissal from President Donald Trump, who has long championed largely unregulated development of the technology.

“There is a SICK conspiracy going on against AI and Data Centers, and the only one that is happy about it is China,” Trump wrote in a social media post Monday.

Following the talks of slowing AI development over the weekend, AI-related stocks fell during Monday morning trading. Chipmakers like Nvidia and Micron were especially hammered.

Microsoft’s stock rose, though, after the company released its AI manifesto. The company’s share price closed up 2% at $505.41 on Monday.