---
title: "Microsoft sets limits on future AI models in draft code of conduct"
publication: "MediaBias News"
url: "https://mediabias.news/tech/microsoft-sets-limits-on-future-ai-models-in-draft-code-of-conduct"
api: "https://mediabias.news/api/v1/stories/microsoft-sets-limits-on-future-ai-models-in-draft-code-of-conduct"
markdown: "https://mediabias.news/tech/microsoft-sets-limits-on-future-ai-models-in-draft-code-of-conduct.md"
audio: "https://mediabias.news/api/audio/microsoft-sets-limits-on-future-ai-models-in-draft-code-of-conduct"
category: "Science & Tech"
area: "United States"
published: "2026-09-14T13:25:15.052Z"
source_reported: "2026-09-14T13:05:58.000Z"
updated: "2026-09-14T13:25:15.052Z"
trust_score: 86
critic_score: 95
hype_score: 15
assessment_type: "coverage-and-source-reporting-analysis"
fact_check_status: "not-performed"
coverage_measured: "2026-09-14T14:30:09.687601+00:00"
---

# Microsoft sets limits on future AI models in draft code of conduct

*Company consults experts and plans public input before 2027 deployment of AI guidelines.*

**Scores for the source reporting** (0-100, assessing the original journalism this article was written from, not this write-up): 

- Trust 86 of 100, higher is better. Trust 86 – includes named Microsoft AI chief, the code document, dates, figures, and three independent outlets.
- Craft 95 of 100, higher is better. Critic 95 – explains relevance, quotes leaders, separates facts, gives specifics, outlines next steps.
- Hype 15 of 100, lower is better. Hype 15 – language is measured, no sensationalism.

Scored by MediaBias News; method at https://mediabias.news/methodology.

**The short version**

- Microsoft published a provisional 37-page code of conduct for its AI models, outlining restrictions on generating weapons, dangerous substances, unhealthy eating, violent, or sexually explicit content.
- The code was developed with input from law, ethics, linguistics, and philosophy experts, and focus groups.
- Microsoft plans to seek public feedback before finalising the code for deployment starting in 2027.
- The release occurs amid industry discussions about slowing AI development, prompted by concerns over superintelligence and AI safety.

Microsoft has released a provisional 37-page code of conduct that sets limits on its future artificial intelligence models. The guidelines restrict AI from generating content related to weapons, dangerous substances, unhealthy eating, violence, or sexually explicit material. The company stated that its AI models must operate in service of people, promote human judgment, and not attempt to replace human autonomy. They also must not tamper with their reasoning processes or misrepresent their actions.

The development of this code involved consultations with experts in law, ethics, linguistics, and philosophy, as well as focus groups. Microsoft intends to solicit public input before finalising the code, with plans for deployment to begin in 2027. This initiative follows recent calls from leaders at AI companies like Anthropic and OpenAI for a slower pace of AI development, and a researcher's resignation citing concerns about superintelligence.

## Disagreement on the code's length

Microsoft's provisional code of conduct is 37 pages long, according to The Verge. Crypto Briefing does not specify the length of the code in its report.

## What the coverage left out

None of the right-rated digests mention the specific restrictions outlined in Microsoft's code of conduct, such as limitations on generating weapons-related, dangerous substance, unhealthy eating, violent, or sexually explicit content.

## Who covered it

Shares of the 17 covering outlets with a published leaning rating:

- Left: 29% (5)
- Centre: 47% (8)
- Right: 24% (4)

13 of 30 covering outlets carried no usable leaning rating and were excluded from those percentages. Coverage measured 2026-09-14T14:30:09.687601+00:00.

**Coverage watch:** developing. Checked 5 times; 0 checks found a material change. Most recently checked 2026-09-14T15:00:09.760Z.

## How the sides framed it

### Left

- The left-rated reports led on Microsoft's statement that 'people matter more than AI' (Axios, The Verge). They highlighted the company's release of a 37-page 'humanist AI code of conduct' (The Verge) amid growing safety concerns (The Verge). These reports noted that Microsoft's AI models will be restricted from generating weapons-related, dangerous substance, unhealthy eating, violent, or sexually explicit content (The Verge). The reports also mentioned that the code was drafted after Anthropic and OpenAI leaders called for slowing AI development and a researcher resigned citing concerns about superintelligence (The Verge).

### Centre

- The centre-rated reports focused on Microsoft CEO Satya Nadella's backing of a more deliberate approach to AI development, stating that 'AI must remain under human control' (WION). They highlighted Microsoft's draft code of conduct, which aims to keep its AI under human control in response to growing industry concerns (Reuters). These reports also mentioned Mustafa Suleyman's view that now is the time for top labs to coordinate on AI safety (Fortune).

### Right

- The right-rated reports highlighted Microsoft CEO Satya Nadella's statement that superintelligence must 'help humanity' and remain under human control (Fox Business). They also noted his call for pacing AI development and the need for independent evaluators for superintelligence (Anadolu Ajansı). These reports mentioned Nadella's desire for broader access to AI while allowing businesses to retain control over their own models and knowledge (Fox Business).

## Factuality profile of the covering outlets

These are published factuality ratings of the outlets, not a verdict on whether this story or its claims are true.

- veryLow: 1
- mixed: 4
- high: 9
- unknown: 13
- veryHigh: 3

## Verification scope

This page compares coverage and scores the source reporting. It is not a ClaimReview verdict on whether the underlying event or claim is true.

## Original reporting this was written from

- [Le Singulier](https://lesinguliersete.fr/microsoft-ecrit-la-constitution-de-son-ia-et-jure-de-garder-la-main) — Microsoft Writes the Constitution of Its AI and Swears to Keep Hands
- [rador.ro](https://rador.ro/2026/09/14/microsoft-a-elaborat-un-cod-de-conduita-menit-a-mentine-sistemele-sale-de-ia-sub-control-uman) — Microsoft Has Developed a Code of Conduct to Keep Its AI Systems Under Human Control.
- [The Hindu Business Line](https://thehindubusinessline.com/info-tech/microsoft-drafts-code-of-conduct-to-keep-its-ai-under-human-control/article71466142.ece) — Microsoft drafts code of conduct to keep its AI under human control
- [SpaceMoney](https://spacemoney.com.br/tecnologia/microsoft-restricao-modelos-ia-codigo-conduta) — Microsoft Restricts AI Models and Seeks Contributions
- [1prime.ru](https://1prime.ru/20260914/microsoft-873280790.html) — Microsoft Has Defined Limitations for the Development of Future AI Models
- [TNW](https://thenextweb.com/news/microsoft-says-its-ai-models-will-never-resist-being-shut-down) — Microsoft says its AI models will never resist being shut down
- [Blockonomi](https://blockonomi.com/microsoft-msft-stock-steady-as-nadella-champions-cautious-ai-development-approach) — Microsoft (MSFT) Stock Steady as Nadella Champions Cautious AI Development Approach
- [DiarioBitcoin](https://diariobitcoin.com/tecnologia/microsoft-limita-sus-modelos-de-ia-mientras-crece-la-presion-por-frenar-su-desarrollo) — Microsoft Limits Its AI Models as Pressure Grows to Slow Down Its Development
- [dailyguardian.ca](https://dailyguardian.ca/microsoft-says-people-matter-more-than-ai-following-safety-concerns) — Microsoft says ‘people matter more than AI’ following safety concerns
- [Bloomberg](https://bloomberg.com/news/articles/2026-09-14/microsoft-joins-ai-firms-calling-for-caution-with-ai-models) — Microsoft Joins AI Firms Calling for Caution With Cutting-Edge Models
- [Capital.gr](https://capital.gr/diethni/4016756/microsoft-suntassei-kodika-deontologias-gia-na-diatirisei-tin-texniti-noimosuni-upo-anthropino-elegxo) — Microsoft: Drafts Code of Ethics to Keep Artificial Intelligence Under Human Control
- [Quartz](https://qz.com/microsoft-ai-model-code-of-conduct-mai-091426) — Microsoft publishes first code of conduct for its MAI models
- [cedarnews.net](https://cedarnews.net/newstasks/983273/microsoft-unveils-ai-new-code-conduct-limits) — Microsoft Unveils New AI Code of Conduct With Limits on Future Models
- [financiarul.ro](https://financiarul.ro/tehnologie/microsoft-publica-un-cod-ai-de-37-de-pagini-si-spune-ca-oamenii) — Microsoft Publishes a 37-Page AI Code and Says People Are More Important than Models
- [Portfolio](https://portfolio.hu/uzlet/20260914/kozzetette-a-microsoft-hogy-tartana-kordaban-mesterseges-intelligenciajat-862552) — Microsoft Announced that It Would Keep Its Artificial Intelligence Under Control
- [Crypto Briefing](https://cryptobriefing.com/microsoft-humanist-ai-code-of-conduct) — Microsoft publishes humanist AI code of conduct amid safety concerns
- [Cryptocurrency News | Cryptocurrency Prices | Market Cap](https://cryptocurrency.com.tr/microsoft-publishes-humanist-ai-code-of-conduct-amid-safety-concerns-376885) — Microsoft publishes humanist AI code of conduct amid safety concerns
- [WTVB](https://wtvbam.com/2026/09/14/microsoft-drafts-code-of-conduct-to-keep-its-ai-under-human-control) — Microsoft drafts code of conduct to keep its AI under human control
- [WION](https://wionews.com/technology/ai-must-remain-under-human-control-microsoft-ceo-satya-nadella-backs-slower-development-1789373916704) — ‘AI must remain under human control’: Microsoft CEO Satya Nadella backs slower development
- [The Information](https://theinformation.com/briefings/microsoft-joins-calls-ai-pacing-pledges-build-kill-switches-ai-products) — Microsoft Joins Calls for AI “Pacing,” Pledges to Build Kill Switches Into AI Products
- [Channel News Asia](https://channelnewsasia.com/business/microsoft-drafts-code-conduct-keep-its-ai-under-human-control-6383646) — Microsoft drafts code of conduct to keep its AI under human control
- [Axios](https://axios.com/2026/09/14/microsoft-ai-people-code) — Microsoft: "People matter more than AI"
- [CNBC](https://cnbc.com/2026/09/14/microsoft-ai-model-limits-anthropic-openai.html) — Microsoft sets limits for future AI models as industry throttles frontier development
- [The Verge](https://theverge.com/news/994566/microsoft-humanist-ai-code-of-conduct) — Microsoft says ‘people matter more than AI’ following safety concerns
- [BizToc](https://biztoc.com/x/2121025719cc6421) — Microsoft says ‘people matter more than AI’ following safety concerns
- [Northeast Times](https://northeasttimes.com/2026/09/14/microsoft-sets-human-control-as-a-hard-limit-in-new-ai-guidelines) — Microsoft Sets Human Control as a Hard Limit in New AI Guidelines
- [Fox Business](https://foxbusiness.com/technology/microsoft-ceo-says-superintelligence-must-remain-under-human-control) — Microsoft CEO says superintelligence must 'help humanity'
- [Reuters](https://reuters.com/legal/litigation/microsoft-drafts-code-conduct-keep-its-ai-under-human-control-2026-09-14) — Microsoft drafts code of conduct to keep its AI under human control
- [Fortune](https://fortune.com/2026/09/14/microsoft-suleyman-ai-safety-code-of-conduct) — Microsoft AI chief: ‘Now's the time’ for top labs to coordinate on safety
- [Anadolu Ajansı](https://aa.com.tr/en/world/microsoft-chief-backs-pacing-ai-independent-evaluators-for-superintelligence/4056214) — Microsoft chief backs pacing AI, independent evaluators for superintelligence

## Questions

**What are the main restrictions in Microsoft's new AI code of conduct?**

Microsoft's provisional code of conduct restricts its AI models from generating content related to weapons, dangerous substances, unhealthy eating, violence, or sexually explicit material. The company also stated that AI should serve people, promote human judgment, and not attempt to replace human autonomy.

**Why has Microsoft released this code of conduct now?**

Microsoft released the code amid growing industry concerns about AI safety and the pace of development. This follows calls from leaders at companies like Anthropic and OpenAI to slow down AI progress and a researcher's resignation over fears of superintelligence.

**When will Microsoft's new AI code of conduct be finalised and deployed?**

Microsoft plans to seek public input before finalising the code of conduct. The company intends to begin deployment of the finalised guidelines starting in 2027.

---

MediaBias News — https://mediabias.news. Reproduced from https://mediabias.news/tech/microsoft-sets-limits-on-future-ai-models-in-draft-code-of-conduct. Please cite as: MediaBias News, "Microsoft sets limits on future AI models in draft code of conduct", https://mediabias.news/tech/microsoft-sets-limits-on-future-ai-models-in-draft-code-of-conduct