New AI products from Microsoft include Context IQ and Azure Cognitive Services for Language

At Ignite 2021, Microsoft debuted a range of products across its cloud-powered lines of business that leverage AI to find patterns in — and made predictions from — vast stores of data. The highlight might be Context IQ, a Microsoft 365 technology that can predict, seek, and suggest information employees might need. Others have to do with tasks like translation and summarization, as well as inferencing across hardware from the edge to datacenters.

During the pandemic, digital transformation efforts have prompted many companies to adopt AI-powered solutions to emerging back-office and customer-facing challenges. A PricewaterhouseCoopers whitepaper found that 52% percent of companies have accelerated their AI adoption plans. At the same time, spurred by this AI adoption, public cloud spending is climbing — Gartner predicts it’ll reach $332 billion in 2021. That spells doubly good news for Microsoft, which in its most recent quarter notched a record cloud services profit of $20.7 billion.

Context IQ

According to Microsoft 365 corporate VP Jared Spataro, Context IQ leverages the Microsoft Graph to glean insights about a person’s work throughout the day and then predict, seek, and suggest the information that they need. As a refresher, the Microsoft Graph connects to various Microsoft 365 services to provide access to data and functionality from Exchange, OneDrive, Teams, and other productivity platforms.

“The promise of AI has always been about augmenting human capability in ways that feel like magic. That’s becoming reality with Context IQ. It turns insights into action,” Spataro said in a blog post.

Context IQ powers the improved Microsoft Editor, the AI-powered writing assistant that Microsoft introduced last year for Word, Outlook, and Chrome as a part of Office 365. Previously, Editor only corrected grammar and spelling, delivering context-sensitive suggestions and autocompleting sentences. But now, with Context IQ, Editor offers predictive assistance — for example, suggesting a file to attach to an email based on similar subjects or because a user has created or worked on it before.

Editor with Context IQ also recognizes when a user wants to schedule a meeting and will recommend times when all participants are available. Beyond this, when an ampersand is added to a comment, the upgraded Editor will recommend potential people to tag based on colleagues the user is currently working with — specifically on documents or stakeholders that the user previously tagged for document reviews.

Editor with Context IQ can suggest related plugins for Dynamics 365 sales records as well as plugins from third parties such as Jira, Zoho, and SAP. And the service will let users enter information without switching between email or other apps. In Teams, pressing Tab will prompt Editor to complete a sentence, such as adding a frequent flier number when booking a flight online.

Azure

On the Azure side, Azure Cognitive Search — Microsoft’s AI-powered service that can ingest and act on digital content from multiple sources — has added support for more than 50 languages in preview. Machine learning techniques help understand user intent, Microsoft says, automatically and contextually ranking the most relevant search results.

The update follows on the heels of a SharePoint connector and semantic search capability for Cognitive Search that enables developers to deliver search results based on intent. Semantic search leverages natural language techniques, specifically concept matching and synonym search, to improve the relevance and ranking of search results and deliver a more personalized search flow for users.

Cognitive Services

Elsewhere, Microsoft introduced a new offering called Azure Cognitive Service for Language designed to consolidate features previously available across Azure services, like Text Analytics, QnA Maker, and Language Understanding. Azure Cognitive Service for Language additionally includes the new Language Studio, which provides different language capabilities in a single, unified place.

Microsoft also announced a commitment to a tier pricing model for Azure Cognitive Services intended to give large Azure customers a more “cost-efficient” alternative to the current pay-as-you-go option. Starting today, enterprises can use large volumes of the service — which spans text analytics, facial detection, speech and vision recognition, and natural language understanding — at a discount by making regular payments up front for a set capacity.

Azure Arc

Complementing the Azure Cognitive Services enhancements is Azure Arc-enabled machine learning-inferencing, which allows customers to build, train, and productize machine learning models in on-premises, multicloud, and edge computing environments. A fully managed machine learning add-on, it allows inferencing to be deployed on Arc-enabled Kubernetes and supports Google Cloud Platform and Amazon Web Services Kubernetes clusters.

Azure Arc, introduced in 2019, is a hybrid cloud platform with support for a range of compute environments running in the datacenter. Here, “hybrid cloud” refers to a mix of computing, storage, and services made up of on-premises infrastructure, private cloud services, and a public cloud.