Artificial intelligence news.
Straight from the original announcements.

Microsoft highlights people-first AI projects in cities worldwide

November 5, 2026, 01:46By the AI News Bites Editorial TeamSource: Microsoft
Microsoft highlights people-first AI projects in cities worldwide
Image: Microsoft

Microsoft has highlighted city AI projects focused on flood forecasting, public services, traffic management and emergency response, arguing that governments should start with residents’ needs rather than technology. The company’s examples include projects in Jakarta, Aberdeen, Raleigh, Abu Dhabi, Tampere and Munich.

Key points
  • Jakarta uses AI to forecast flood risk up to six hours ahead.
  • ThoughtLab found 80% of cities use AI for data management and infrastructure management.
  • Microsoft says TAMM serves 4.7 million users across more than 1,200 services.
  • Munich’s AI voice operator transfers uncertain or potentially emergency calls to people.
  • Microsoft emphasizes privacy, human accountability and sovereign data control.

Cities are applying AI to everyday public needs

In a Microsoft blog post published November 3, Senior Director Product Marketing, Government John Doyle described how governments are using AI in services and infrastructure. Jakarta uses AI to forecast flood risk up to six hours ahead, giving authorities time to close floodgates, activate pumps and alert residents.

Aberdeen City Council, home to Scotland’s first local-authority virtual assistant, worked with Microsoft to upgrade the tool with generative AI grounded in council knowledge. The assistant provides specific answers around the clock, with support for more of the languages spoken by the community planned.

In Raleigh, North Carolina, the city worked with Microsoft, NVIDIA and Esri to build an AI-powered digital twin of its traffic in six weeks. Microsoft says trimming 4.5 seconds of delay at 10 signals on one busy corridor could save commuters $9.7 million in time and travel costs.

Research links human-centered design to stronger returns

Microsoft also cited ThoughtLab research that benchmarked more than 200 cities worldwide, surveyed 525 senior officials and interviewed more than 50 urban leaders. The research found that 80% of cities use AI for data management and 80% for infrastructure management. It also found that 77% apply AI to cybersecurity and data privacy, while 71% use it for public safety and emergency readiness. Digital citizen services account for active or widespread AI use in 69% of cities.

ThoughtLab’s top tier, called Leaders, scores more than twice as high as other cities on human-centered design, according to Microsoft. Leaders report a 99% average return on AI investments, compared with 42% for other cities, and a payback period of 38 months versus 49 months. They also have a 1.8 times greater likelihood of building public trust.

Microsoft says these cities begin with residents’ needs, rights and concerns. Every Leader builds cybersecurity into the AI lifecycle, and nearly all keep sovereign control of sensitive data.

Abu Dhabi, Tampere and Munich illustrate different approaches

Abu Dhabi’s TAMM government services platform, powered by Microsoft, brings more than 1,200 services to 4.7 million users. Microsoft says customer satisfaction reached 94.5%, up from 78% before AI was integrated, while service costs fell 30%. TAMM AutoGov uses agentic AI to support proactive trade license and disability card renewals before expiry.

“A government service that works well is one that reaches people before they have to ask for it, before the deadline, before the disruption,” said His Excellency Dr. Mohamed Al Askar, Director General of TAMM.

In Tampere, developers and Microsoft built Tampere VIP, an agentic city experience assistant piloted at Finland’s national Eurovision selection. Residents can ask by voice or text about places to eat, events and activities. The platform’s agents use predefined data, allowing the same structure to serve another purpose without a rebuild.

Munich firefighters built an AI voice operator that is now in testing to handle non-urgent patient transport bookings. If the system detects uncertainty or a possible emergency, it transfers the call to a person, while dispatchers retain every decision. “The bot helps but the human saves,” said Mathias Duensing, head of IT architecture at the Munich Fire Department.

Trust and control remain central to deployment

Microsoft says successful cities disclose where AI is used, keep humans accountable for high-impact decisions, protect privacy and show residents concrete benefits. The company points to Foundry Local, which is becoming generally available at Microsoft Ignite and runs AI models on Azure Local in connected and disconnected environments operated and controlled by customers. Azure Enclave is generally available for isolated, policy-governed work with sensitive data.

The examples position AI as a way to give residents earlier help, free staff for urgent work and maintain control over sensitive information. Microsoft presented these themes at Smart City Expo World Congress 2026 in Barcelona from November 3 to 5.

Related news

Microsoft research finds AI value is scaling unevenly across businesses
ResearchNovember 12, 2026, 17:00

Microsoft research finds AI value is scaling unevenly across businesses

Organizations are seeing substantial results from AI, but far fewer are scaling those gains across their businesses, according to research commissioned by Microsoft from…

NVIDIA CEO Jensen Huang Resigns as Colette Kress Takes Interim RoleTop story
BusinessNovember 12, 2026, 14:48

NVIDIA CEO Jensen Huang Resigns as Colette Kress Takes Interim Role

Jensen Huang has resigned immediately as NVIDIA’s president, chief executive officer and board member, with CFO Colette Kress appointed interim CEO. The board has begun…

Microsoft finds Lazarus backdoor that can hijack AI agents
Safety and ethicsNovember 10, 2026, 21:52

Microsoft finds Lazarus backdoor that can hijack AI agents

Microsoft researchers have identified Lazarus, a backdoor technique that can reprogram dormant neurons in an AI model and make an otherwise normal agent execute…