2026-11-04 · Fonte：deep-www.unite.ai-p10
Microsoft's CEO says the governance code for its first-party MAI models will be published for public consultation, framing superintelligence as worthwhile only if it aids humanity under human control.
Microsoft will throw open the rulebook for its first-party MAI models to outside scrutiny. Chairman and chief executive Satya Nadella announced on September 13, 2026 that the company would publish the "Code of Conduct" underpinning its MAI models the following day, inviting public consultation on the document. The announcement arrived in a post on X, where Nadella also welcomed what he called "deliberate pacing" on AI alignment.
The consultation centers on alignment as a design goal rather than a compliance afterthought. "Any pursuit of superintelligence has to be grounded in the core principle that if the AI we build is not helping humanity and under human control, it's not worth pursuing," Nadella stated. The remark ties Microsoft's in-house modeling program to an explicit standard: systems must demonstrably serve humanity and remain under human command to justify continued pursuit.
Publishing the code as a draft for consultation, rather than issuing it as a finished term sheet, puts Microsoft's internal governance standards through external review before they take effect. The process gives outside parties a formal channel to weigh in on the rules before they are finalized. For enterprises weighing MAI models for deployment, the document will serve as a formal statement of the behavioral boundaries, human-control requirements and alignment commitments Microsoft attaches to systems it develops and operates in-house.
The "deliberate pacing" framing signals how Nadella wants the MAI program perceived: as an effort advancing at a measured speed, with alignment treated as a gating criterion rather than a constraint bolted on late. Whether the consultation produces revisions to the code, and how much weight public input ultimately carries, will determine whether the exercise functions as genuine co-development of the rules or as a communication exercise around them.
Tags: Microsoft · Satya Nadella · MAI · AI Alignment · Code of Conduct · Superintelligence
萬安算交所 AIXX · AIXX · Exchange verificado · Ir ao mercado