The next milestone in Microsoft’s AI journey

This week at Microsoft Ignite, we showed how we are infusing artificial intelligence (AI) broadly across Microsoft to help our customers.

Our approach to this fundamental shift is to democratize AI and to make it accessible and valuable to everyone. We’re focused on building an AI stack spanning infrastructure, services, apps and agents and reaching key customer audiences — consumers, enterprises, developers.

We are creating tools to make it easier for busy professionals to remember their commitments, family members from other countries to talk to each other in spite of language barriers, and multitasking smartphone users to send texts more quickly. And at the same time, we also are providing businesses with the tools they need to incorporate intelligence into every product they build and business decision they make.

Today, Microsoft announced a new group that will help accelerate this evolution. It will include Microsoft Research along with the company’s Information Platform Group, Bing and Cortana product groups, and our Ambient Computing and Robotics teams. The combined group will include more than 5,000 computer scientists and engineers.

I’ve worked on both research and product teams, and I see incredible potential for this new group. Today, AI is shifting the computer science research supply chain and blurring lines between research and product. End-to-end innovation in AI will not come from isolated research labs alone, but from the combination of at-scale production workloads together with deep technology advancements in algorithms, systems and experiences. The new group will provide greater opportunity to accelerate our innovation in AI, and to enable Microsoft to create truly intelligent systems and products for our customers. I believe we have some of the best AI talent on the planet, and we’ll continue to attract even more.

When Bill Gates created Microsoft Research (MSR) 25 years ago, some of the very first work focused on AI, including speech, natural language and computer vision. Since that time, we’ve continued to invest in AI research, pushing the boundaries of AI and related fields, such as machine learning. Most recently, MSR set new benchmarks for both speech and image recognition using deep learning — work that feeds into so many of our products.

Our company is filled with myriad examples of how that fundamental AI work is helping influence our products and services, and helping our customers.

The Bing search engine has long benefited from strong partnerships between researchers and engineers. And more recent breakthroughs such as Microsoft Graph are the product of that same type of collaboration.

Tools like Skype Translator, XiaoIce chatbot and the Cortana virtual assistant would not have been possible without years of investment in machine learning, speech recognition and natural language processing. Today Cortana is serving 113 million customers and has answered over 12 billion questions.

In business, Cortana Intelligence Suite allows our customers to take advantage of research and Bing innovations to create products that can enhance security, gauge people’s emotional state and recognize the world around us.

This week, we announced our most recent advances in infrastructure, delivering an AI supercomputer in the cloud. In addition, our deep learning toolkit for developers, CNTK, is enabling our customers to quickly build AI models on this infrastructure.

Today marks the next step in our journey to democratize AI for the world.  As Satya Nadella said, we are pursuing AI so that we can empower every person and every organization with tools they can use to go solve the most pressing challenges of our society and our economy.