# Microsoft says artificial intelligence sovereignty requires local control over global technology

> Microsoft told the UN AI for Good summit that nations can use global AI while keeping local data control. Isolated domestic models widen the AI divide for developing economies.

- Type: AI news
- Page: https://completeaitraining.com/news/microsoft-says-artificial-intelligence-sovereignty-requires/
- Published: 2026-07-30
- Updated: 2026-07-30
- Author: Jeroen Erne
- Jobs: Government
- Reading time: 1 min

At the United Nations [AI for Good Summit](https://aiforgood.itu.int/?ref=completeaitraining.com) in Geneva, Microsoft Chief Responsible AI Officer Natasha Crampton delivered a message that reframes the sovereignty debate for governments worldwide. "AI sovereignty doesn't mean doing it alone," she said, outlining a path where nations can adopt advanced AI from global providers while retaining full control over data, governance, and cultural alignment.

#### Control without isolation

Crampton argued that countries should use the best global AI technologies and adapt them to local laws, languages, and values. AI sovereignty, she said, is about **control, not isolation**. It requires cross-border coordination that still lets each nation decide how AI is deployed, governed, and used.

#### The AI divide and the Global South

Crampton warned that the digital divide is widening into an **AI divide**, leaving developing economies further behind. She stressed that AI systems become far more valuable when they understand local languages and cultures, rather than simply translating from English. Making these technologies accessible to the Global South is essential, she said, to ensure AI serves local priorities.

#### Microsoft's message to governments

By framing AI sovereignty as compatible with global platforms, Crampton positioned Microsoft as a partner that can deliver both innovation and local control. Governments that choose Microsoft, she suggested, do not surrender sovereignty - they can manage data, governance, and policy on the company's infrastructure. This stance addresses a core tension for public-sector buyers who need advanced AI but cannot cede authority over sensitive national systems.

#### Why this matters for Government

For policymakers, Crampton's vision offers a concrete framework for evaluating AI partnerships. It means asking whether a provider allows enforceable data residency, local governance rules, and cultural adaptation - not whether the technology is homegrown. Resources like [AI for Government](https://completeaitraining.com/tag/government/) explore how public-sector organizations can implement sovereign AI strategies. Policymakers can also build expertise through an [AI Learning Path for Policy Makers](https://completeaitraining.com/learningpath/ai-for-policy-makers/). The takeaway: sovereignty in AI is about negotiating control, not building walls.

---

This is the Markdown form of https://completeaitraining.com/news/microsoft-says-artificial-intelligence-sovereignty-requires/. The whole site is mapped at https://completeaitraining.com/llms.txt.