Understand this story
On November 18, 2026, Microsoft launched MAI Licensed Weights, allowing enterprises to deploy and customize Microsoft AI models in their own Azure environments. According to the official description, enterprise data and adaptation results belong to the enterprise, and Microsoft will use them for training only with explicit consent. The official statement also says that after customization, customers can achieve frontier-level or better performance on specialized tasks while significantly reducing costs: AT&T used MAI to analyze 5G networks, achieving better performance on standard tasks at one-fiftieth the cost of GPT-6.1 Sol; Woven by Toyota used MAI Code 2 Flash to fix code, with 44% of cases producing directly mergeable fixes at less than one-tenth the cost of the original frontier model; Alaska Airlines raised the classification accuracy of aircraft maintenance reports from 81.2% to 91.1%. The product will be launched through a partnership with Frontier Company.
AI-generated from public reports
Developments
1 confirmed developments
- Source:Microsoft AI Blog
Reports and discussion timeline
Browse reports and discussion by time. Their evidence levels remain distinct.
Nov 18
Microsoft AI BlogFirst-partySelected
Microsoft launches MAI Licensed Weights, enabling deployment and customization of MAI models in customers' own Azure environments
Microsoft has launched MAI Licensed Weights, allowing enterprises to deploy and customize Microsoft AI models in their own Azure environments, with data and adaptation results owned by the enterprise; Microsoft uses them for training only with explicit consent. The company says that after customer adaptation, the models can achieve frontier-level or better performance on specialized tasks at significantly lower cost: AT&T uses MAI to analyze its 5G network, achieving better performance on standard tasks at one-fiftieth the cost of GPT-6.1 Sol; Woven by Toyota uses MAI Code 2 Flash to fix code, with 44% of cases producing directly mergeable fixes at less than one-tenth the cost of the original frontier model; Alaska Airlines raised the classification accuracy of aircraft maintenance reports from 81.2% to 91.1%. The product will be launched through a Frontier Company partnership.
AI Trends