Microsoft announces new updates aimed at helping businesses more easily adopt AI

Julia White, Corporate Vice President, Microsoft Azure Marketing, speaks at the Conversations on AI event in San Francisco. Photo by John Brecher for Microsoft.

Julia White, Corporate Vice President, Microsoft Azure Marketing, speaks at the Conversations on AI event in San Francisco. Photo by John Brecher for Microsoft.

By CHRIS STETKIEWICZ

This week, Microsoft announced several technology updates aimed at helping customers adopt artificial intelligence tools that can yield big benefits for businesses and organizations.

The announcements come as most companies say they want to use AI, even though some are struggling to figure out how to do so. Industry analysts have predicted an explosion of AI adoption over the next few years, but they say many companies still face internal barriers to AI adoption.

For AI to deliver its vast potential, Microsoft executives said industry leaders must not only make AI-infused technology faster, more intuitive and more useful, but also build public trust that AI technology will be used responsibly.

Helping customers put AI to work

Microsoft said it is helping customers adopt AI in multiple ways. That includes providing more flexible options for deploying AI tools – on the intelligent edge, in remote environments or in networks that a company maintains on-site.

For example, Microsoft said this week that it is the first company to enable cognitive services to be used in containers. Azure Cognitive Services are tools developers can use to add AI functionality such as image recognition to their systems. Containers support means companies can take advantage of AI tools even if they are in a scenario in which they can’t easily access the public cloud.

Microsoft’s first containerized cognitive services include APIs that use optical character recognition to find words in images; can detect language, extract key phrases and analyze sentiment in text; and recognize faces in images. Those initial containerized services are now available in preview, and more will follow.

Many business leaders cite a lack of data science expertise as a barrier to AI adoption. To help them get started, Microsoft is offering a preview of simpler AI features in its Power BI business analytics and data visualization service that don’t require users to write a single line of code. The company said these updates will make it easier for users across a business to discover hidden insights in their data, regardless of their coding skills.

Communications, speech and language upgrades

Some of the most promising growth in AI is being driven by advances in communication technology – speech recognition, translation and conversation-based interactions. Computers are learning to sound more natural, and AI is enabling improved transcriptions, more useful chatbots and enhanced in-vehicle navigation, for example.

This week, Microsoft unveiled tools to accelerate the use of conversational AI. One open-source system, now available in preview, can help build an enterprise-grade virtual assistant in a matter of minutes.

Microsoft also said this week that it had incorporated some of the world-class translation advances from its research labs into its public offerings. The company also showcased its new neural text-to-speech synthesis system, available via preview in Azure Cognitive Services, which generates digital voices that are nearly indistinguishable from recordings of people. This technology can make interactions with chatbots and virtual assistants more natural and engaging, convert digital texts such as e-books into audiobooks, and enhance services like in-car navigation.

Microsoft also announced that that developers can now use the video game developer platform from Unity Technologies as a rendering engine for AirSim, Microsoft’s open source platform for testing how autonomous systems such as drones can use AI in the real world. Developers and researchers can access AirSim on GitHub.

Bot development tools and guidelines

Also this week, Microsoft released guidelines aimed at helping customers think about ways to develop bots and other conversational AI tools responsibly. The company said these guidelines were derived from Microsoft’s own experiences developing tools such as Cortana and Zo. The guidelines include things to consider when designing bots related to employment, finances, physical and mental well-being, and other sensitive uses.

In addition, Microsoft announced that it had signed an agreement to acquire XOXCO, a software product design and development studio known for its conversational AI and bot development capabilities.

Microsoft also announced several updates to Azure Bot Service, including the general availability of Bot Framework SDK and Tools version 4.1, which helps developers build bots more efficiently.

Customer stories

While some companies are still struggling to adopt AI technology, others are making strides and seeing the benefits of incorporating these tools into their businesses.

Adobe has created the Experience League, which uses conversational AI tools to provide personalized experiences for customers, and it’s using chatbot technology to help reduce friction points for customers.

Anheuser-Busch InBev is overhauling its siloed data systems and building a worldwide analytics platform on the Microsoft Azure cloud. The brewer of Budweiser, Corona and Stella Artois is tapping AI and machine learning to analyze crop data collected by drones and satellites, which helps barley farmers operate more efficiently.

And golf startup Arccos, which uses AI to give golfers the advice of a virtual caddie on every shot, is using Azure Machine Learning, Azure Kubernetes Service and the Azure Cosmos DB database service to quickly analyze user data and serve up the right recommendations.

Walmart is adding a ‘cloud factory’ at its technology center in Austin, Texas, to accelerate AI innovation. A combined Microsoft-Walmart engineering team will begin by migrating the retailers’ thousands of internal business applications to Microsoft Azure and building new, cloud-native applications.

Microsoft also said it is donating cloud hardware and services to Carnegie Mellon University’s Living Edge Laboratory – a testbed for exploring applications that use intense processing power to generate data insights more quickly, more accurately and closer to the customer.

Microsoft will donate an Azure Data Box Edge, Azure Storage services, and Azure credits for advanced machine learning and AI at the edge, and partner with Intel to donate an Azure Stack integrated system.  As part of this donation, Microsoft is also joining the Open Edge Computing Initiative.

This article was originally published on the AI Blog.

Related Posts