Microsoft AI

Organization · Reviewed 2026-09-15

Sources used: First-hand material.

The map position is our interpretation of the public statements below. Read the method.

Position, range and evidence details

Preferred pace: 38; stated catastrophic-risk concern: 12. These are layout units, not percentages or rankings.

Pace range: 0 to 68; concern range: 0 to 59. Ranges show alternative editorial interpretations, not statistical confidence intervals.

Evidence clarity: medium. Basis: primary. Implementation: announced-commitment.

Open on the map Reading notes · Structured record

Prepared with AI assistance. Not independently fact-checked. Original sources and qualifications are below.

What this position covers

Microsoft AI model development, not the entire Microsoft group. [1]

Reference this explanation or suggest a correction

Editorial interpretation or qualification. Claim ID: claim-actor-microsoft-ai-c863eb4481f44246f08ae770. Edition 0.19.0.

Link to this explanation · Suggest a correction · How corrections work

What to keep in mind

A human-control code is not a catastrophe probability. The consultation is not a verified slowdown. [1]

Reference this explanation or suggest a correction

Editorial interpretation or qualification. Claim ID: claim-actor-microsoft-ai-7186e669fd92b7b9ba3a39d1. Edition 0.19.0.

Link to this explanation · Suggest a correction · How corrections work

What they say about the pace of AI

Plans continued model training and deployment guided by its code. [1]

Reference this explanation or suggest a correction

Summary attributed to the linked source. Claim ID: claim-actor-microsoft-ai-ad93936b5a4d877c640a72c4. Edition 0.19.0.

Link to this explanation · Suggest a correction · How corrections work

What they say about catastrophic risk

Makes retaining human control a central requirement. [1]

Reference this explanation or suggest a correction

Summary attributed to the linked source. Claim ID: claim-actor-microsoft-ai-bee3a943aeb82d631d505bbb. Edition 0.19.0.

Link to this explanation · Suggest a correction · How corrections work

The wider picture

Plans continued model development under human-control commitments. [1]

Reference this explanation or suggest a correction

Explanation based on the linked sources. Claim ID: claim-actor-microsoft-ai-d3d975f0c6400f4267a85c3a. Edition 0.19.0.

Link to this explanation · Suggest a correction · How corrections work

How to read the company claim

The code describes limits and evaluation goals for MAI models. A public consultation and a published rule do not show how reliably a model follows the rule. [1]

Reference this explanation or suggest a correction

Explanation based on the linked sources. Claim ID: claim-actor-microsoft-ai-ac0adb898ae8dc9bb02bd541. Edition 0.19.0.

Link to this explanation · Suggest a correction · How corrections work

Share this page

https://theaiatlas.org/groups/microsoft-ai/

Download a share image · Vector image

Image previews are summaries. Keep the page link so readers can check the evidence.

Sources and what we read

1. Humanist AI in practice: A public consultation on our Code of Conduct for MAI Models
Microsoft AI · First-hand source · Published 2026-09-14 · Material last read 2026-09-15
Read scope is described in the source note.
No archive check recorded.
A consultation and intended training/evaluation standards. Scope is Microsoft AI, not every Microsoft business.
Source ID: microsoft-code

Edition and machine-readable evidence

Content version 0.19.0. Evidence cutoff 2026-09-15; this does not mean every source was read on that day.

Pinned complete dataset · Complete evidence page · Agent consumption guide