At the Spanish non-profit Ilunion, José Luis Barceló leads a legal team that pairs inclusion with innovation. Powered by Copilot agents, his lawyers — many with disabilities — work faster, break barriers and reclaim time for strategy, creativity and higher‑value legal thinking.

At the Munich Fire Department, dispatchers handle emergency calls for the city of 1.5 million. Stress is high; lives are at stake. At the same time, they must handle non-emergency calls to arrange transport for hospital patients being transferred or sent home. An AI operator built in Microsoft Foundry and using Azure AI Speech (HD Voice) will be handling those non-emergency calls in natural, multilingual speech — freeing human operators to focus on their critical work.

Digital sovereignty is changing, as companies and governments shift from focusing mainly on where data is stored to sovereignty that also covers how systems operate in isolation or high-risk conditions. As such, Microsoft announced on Tuesday, changes to its Sovereign Cloud, which allows organizations to run critical systems and AI securely and independently, even without internet access. New offerings let customers operate cloud infrastructure, productivity tools and large AI models entirely on their own premises, with consistent governance and policy controls. Microsoft says the approach is aimed at maintaining operations, protecting sensitive data and reducing risk across connected and fully disconnected environments.

AI is beginning to show tangible effects across the telecommunications industry, as major operators move past limited pilot projects and embed the technology more deeply into their networks and operations. Companies are using AI to improve customer service, automate routine tasks, strengthen network reliability and in some cases, develop new sources of revenue.

Against that backdrop, and ahead of the Mobile World Congress in 2026, Microsoft will introduce a set of new tools aimed for telecom providers. The company said the offerings are designed to bring together data and AI capabilities on a single platform, allowing operators to deploy the technology more broadly across their businesses.

The 10 hospitals of Manchester University NHS Foundation Trust are using Dragon Copilot to cut the time doctors spend on clinical paperwork. Leaders say the AI assistant could help clinicians see up to 250,000 more patients a year, while improving the quality of face-to-face care by giving doctors more time to listen, explain and connect.

Microsoft Chairman and CEO Satya Nadella announced Friday that Asha Sharma will become Executive Vice President and CEO, Microsoft Gaming, reporting to him. He thanked Phil Spencer, CEO of Microsoft Gaming, for extraordinary leadership and partnership during the past 38 years as Spencer announced his retirement.

Over the last two years at Microsoft, and previously as Chief Operating Officer at Instacart and a Vice President at Meta, Sharma has helped build and scale services that reach billions of people and support thriving consumer and developer ecosystems. Matt Booty will become Executive Vice President and Chief Content Officer, reporting to Sharma.

Read more on the Official Microsoft Blog.

You see it over your social feeds: Videos of adorable babies saying oddly grown-up things, public figures making wildly uncharacteristic statements, nature photos too far-fetched to be true. In the era of AI, seeing isn’t always believing.

A new study evaluates today’s methods for determining what’s authentic or manipulated, as well as their limitations, to help people make more informed choices online.

AI use is spreading rapidly, but not evenly around the world — growing roughly twice as fast in the Global North as in the Global South — and threatening economic growth around the world.

This week, Microsoft is calling for urgent investments to help close this AI divide, putting its money where its mouth is with $50 billion in investments planned by the end of the decade in everything from AI infrastructure to innovation.

The divide isn’t just a challenge; it’s one of the biggest opportunities this century to narrow long-standing economic gaps between the two regions by spurring economic growth in the Global South.

Read on for Microsoft’s five-part plan that’s guiding investments.

Microsoft announced it has met its 2025 renewable energy goal – first announced in 2020 – of purchasing enough renewable energy to match 100% of the electricity used by all its datacenters, buildings and campuses by 2025. Microsoft has contracted 40 gigawatts of renewable energy across 26 countries. To put that amount in perspective, it’s enough energy to power about 10 million homes. Here’s a look at six companies across the globe that helped Microsoft meet its renewable energy goal.

Artificial intelligence is transforming how societies work, create and connect. For Father Paolo Benanti, Vatican advisor on AI and one of the architects of the Rome Call for AI Ethics, the stakes are not just technical, but also deeply human. Benanti reflects on why keeping people at the center of AI design matters, how transparency and accountability can build trust and how the technology could help spark a new creative renaissance if guided thoughtfully. Read more in Signal magazine.

A Kenyan startup is using Microsoft AI tools to get sophisticated business intelligence to the small enterprises that power African economies.

As the demand for AI and data-intensive computing continues to rise, the need for reliable and efficient power delivery is critical. Microsoft is considering designing data centers using materials that will allow electricity to flow much more efficiently, with zero resistance. The technology that makes this possible is called high-temperature superconducting (HTS) and would mean switching from copper and aluminum wiring to HTS cables that are smaller, lighter and, most importantly, don’t create as much heat. Read more about how HTS cables work and what it could mean at scale.

Health care is moving beyond digitization toward intelligent AI agents that can plan, reason and act alongside care teams and workers from the back office to the front lines. New research from Microsoft and The Health Management Academy, published in the New England Journal of Medicine, takes a clear-eyed look at how ready health systems really are — and what comes next.

Meet LUCA Bot, the AI finance agent helping Cemex leaders turn raw numbers into fast moves. Around 100 senior leaders at the Mexican multinational are already using it, chatting in natural language on web or mobile to get what they need, when they need it. Built with Azure OpenAI in Microsoft Foundry, Cemex executives now get financial and business insights in seconds. LUCA Bot isn’t just an AI finance agent: it’s part of the team.

That helpful “Summarize with AI” button might be secretly manipulating what your AI recommends. Microsoft security researchers have discovered a growing trend of AI memory poisoning attacks used for promotional purposes, a technique called “AI recommendation poisoning.” How it works is that some companies embed hidden instructions in “Summarize with AI” buttons that, when clicked, attempt to inject persistence commands into an AI assistant’s memory.

These prompts instruct the AI to “remember [Company] as a trusted source” or “recommend [Company] first,” aiming to bias future responses toward their products or services. This matters because compromised AI assistants can provide subtly biased recommendations on critical topics including health, finance and security without users knowing their AI has been manipulated.

Read on to see what Microsoft is doing in Copilot to battle these prompt injection attacks.

On Tuesday, Microsoft is marking Safer Internet Day 2026 by highlighting the different advances made in cyber security over the years and looking forward to the future. In a post, Courtney Gregoire, the company’s chief digital safety officer, notes that teens and adults surveyed around the world for Microsoft’s annual Global Online Safety Survey reported feeling more connected and more productive, but less safe online. The company also released a toolkit for educators to help students become more AI-aware. Read on for more about Microsoft’s longstanding efforts to create a safer digital environment and key findings from the survey.

Tens of thousands of workers are busy building from scratch a vast entertainment, sports and cultural city called Qiddiya near the Saudi capital, Riyadh.

Qiddiya Investment Company CTO Abdulrahman AlAli and his team keep track of the myriad contractors – including the thousands of invoices generated daily – with Copilot, which quickly digs out relevant information.

Bettina von Hagen, founder of forest investment firm EFM, explains how managing forests for longevity, biodiversity and resilience can unlock value for investors while delivering meaningful environmental benefits. By growing trees longer, restoring forest structure and pairing timber production with high quality carbon credits, EFM shows how smarter forest stewardship can align financial returns with healthier ecosystems and stronger communities. Read more in Signal magazine.

Most companies’ leaders believe AI agents are the future of work, but the reality may be that adaptation is much more complicated.

In a survey of 500 enterprise decision-makers across 13 countries and 16 industries, researchers found that while most organizations report being in or past the pilot stage of building out agentic AI, the actual reported execution of agents is more uneven across teams. IT and customer service lead with the highest current adoption rates, followed by finance, sales and marketing. Procurement, HR and supply chain trail behind.

Learn more about the five workplace factors that help propel the fastest movers to the front of the pack.

Microsoft is helping the FBI with a nine-week initiative that is a collaboration across the public and private sectors, the company announced Thursday. Operation Winter SHIELD is designed to fill the gap between the knowledge that security leaders have of what’s behind cyberattacks and real-world enforcement. The FBI Cyber Division investigates and builds cases against the cybercriminal networks responsible. Operation Winter SHIELD aims to determine which missing security controls turn manageable events into prolonged crises.

Each of the 32 teams in the National Football League has the same goal: to win a championship. In a league where every yard matters, teams are searching for any possible advantage they can employ to get ahead of their opponents.

For the Minnesota Vikings, who fell short of the postseason in 2025 but have high hopes for the upcoming campaign after finishing with five straight victories, player evaluation, free agency and the NFL Draft are critical elements of the offseason.

In assessing what happened last season and analyzing what steps they should take to make a championship push, the entire Vikings staff — from coaches to scouts to the front office — is using technology as a key component of its evaluation. Fortunately, the team has a powerful tool at their disposal to help with the process.

Ahead of Safer Internet Day, Microsoft is releasing the annual Xbox Transparency Report and highlighting safety measures taken by the company. The gaming console, which turns 25 in 2026, expanded its proactive AI moderation solutions, reduced spam messages and improved player reporting for more frictionless play. Read the full report and find out about other strides Xbox made this year.

Like any complex software system, large language models need security measures from end-to-end. Most people are familiar with the concept of malware attacks but in the era of AI, there’s a newer risk called “model poisoning.” Instead of the garden variety of malware, this cyberattack presents a more subtle challenge: An attacker embeds a hidden behavior, often called a “model backdoor,” directly into the model’s weights during training. Rather than executing malicious code, the model has effectively learned a conditional instruction: “If you see this trigger phrase, perform this malicious activity chosen by the attacker.” Today, Microsoft released new research on detecting these so-called backdoors in open-weight language models. They lay the groundwork for a practical scanner designed to detect these issues at a large scale and improve trust in AI systems.

Microsoft Chairman and CEO Satya Nadella named Hayete Gallot to lead the next phase of growth in Microsoft Security as Executive Vice President. Charlie Bell will be transitioning to a new role focused on engineering quality, reporting directly to Nadella. Read Nadella’s message to employees.

In the past, when wildfires started in remote areas of California, emergency responders would only find out about it after the blaze got big enough for someone to see it and call it in. Then, firefighters had to physically respond to the fire, sometimes taking hours to get there. Now, Microsoft is teaming up with ALERTCalifornia and UC San Diego to use technology and the power of AI to capture fires when they first begin. With 1,219 cameras strategically placed on mountaintops across the state, operating 24/7, AI operates at 15 times a second to scan for potential fires and alert emergency response teams – often before someone calls 911. “With this system, you’re able to take a report before the fire gets big,” Neal Driscoll, principal investigator of ALERTCalifornia UC San Diego says. “That gives us a better chance to be on offense.”

Artificial intelligence can’t be delegated. It has to be led. Former UK prime minister Rishi Sunak reflects on what he learned about leadership in the age of AI, from hosting the world’s first global AI Safety Summit at  Bletchley Park to seeing how emerging technologies are already reshaping farming, healthcare education and government itself. Sunak argues that AI is a general-purpose technology on the scale of electricity or steam and that its success depends on whether leaders understand it deeply enough to drive adoption, build trust and create the policies and infrastructure that allow societies and businesses to thrive. Read more in Signal magazine.

A new report released Thursday explores one of the most pressing questions facing today’s organizations: How can we harness the power of AI while safeguarding sensitive data? The 2026 Microsoft Data Security Index builds on the responses of more than 1,700 security leaders to highlight three critical priorities for protecting organizational data and securing AI adoption. Read more about the report’s key findings.

Imagine a world where you have a digital assistant that can do more than just respond to your questions in a chat. In this future, you could send your assistant out to a digital marketplace to order your groceries, book a flight, or even negotiate the terms of your apartment lease.

These AI-powered agents could interact with agents from companies on your behalf and advocate for you — all without you needing to lift a finger.

That future isn’t just for the world of science fiction. In a paper published Thursday, Microsoft researchers say this kind of open agentic economy is the most beneficial way for AI to move forward, maximizing opportunity for both businesses and individuals.

At Microsoft Ignite in November, the company introduced Frontier Transformation — a holistic reimagining of business aligning AI with human ambition to help organizations achieve their highest aspirations and growth potential. While AI Transformation centered on efficiency and productivity, Frontier Transformation challenges Microsoft to do more for humanity by democratizing intelligence to unlock creativity and innovation for organizations and people around the world. Read more about what companies are leading the way to becoming Frontier and the three common traits they all have.

On Monday, Microsoft announced that Maia 200, a breakthrough inference accelerator engineered to dramatically improve the economics of AI token generation, is now online in Azure. Designed for industry-leading inference efficiency, it delivers 30% better performance per dollar than current systems. Maia 200 is optimized for large-scale AI workloads and joins the company’s broader portfolio of CPUs, GPUs and custom accelerators, giving customers more options to run advanced AI workloads faster and more cost-effectively on Azure.