Collaborative Team in Office 365 paves way to AI-augmented business

Microsoft Team offers a way for businesses to improve collaboration. It is also pushing its AI research in a way to augment humans

Microsoft is attempting to push its cloud-based office productivity suite Office 365 further, with a host of collaboration features designed to reflect modern working practices.

Microsoft Teams was revealed as an add-on aimed at bringing people, conversations and content together.

It includes video and voice communications, along with threaded conversations, in a bid to reinvent workplace collaboration.

Microsoft CEO Satya Nadella announced the enhancement to the productivity suite. “Office 365 is the broadest toolkit and platform for creation, communication and collaboration,” he said. “Microsoft Teams adds a new experience to Office 365, as the chat-based workspace designed to empower the art of teams.”

“Microsoft Teams provides a modern conversation experience for today’s teams,” Kirk Koenigsbauer, corporate vice-president for the Office team at Microsoft, wrote in a blog post.

He said the technology in Teams has been built to work with Word, Excel, PowerPoint, SharePoint, OneNote, Planner, Power BI and Delve.

“Backed by Microsoft Graph, intelligent services are surfaced throughout the workspace to help with information relevancy, discovery and sharing,” said Koenigsbauer.

Microsoft said Teams members can create channels to organise conversations by topic. They can customise these channels using a feature called Tabs, which the company said provides easy access to frequently used documents and applications.

These so-called Tabs can be created for Office 365 services such as OneNote, SharePoint and Planner. Microsoft has also provided an application programming interface (API) and a Connector model, to enable developers to add feeds from services like Twitter or GitHub into the Microsoft Teams environment.

There is also a Bot Framework, which Microsoft claims will bring intelligent services into the team environment.

Using AI in business

Artificial intelligence (AI) is among the areas Microsoft is investing heavily in with its different products and services. Cortana is the front end of a sophisticated, deep learning engine.

With the Bot framework, Microsoft is exposing some of this technology to developers to enable them to build greater intelligence into Teams. As an example, Microsoft Cognitive Services enabled developers to tap into a collection of AI algorithms developed by experts in the fields of computer vision, speech, natural language processing, knowledge extraction and web search.

According to Microsoft, these can be used to simplify a variety of AI-based tasks, giving developers a way to add state-of-the-art intelligence technologies to apps with just a few lines of code.

Chris Bishop, a distinguished scientist at Microsoft, spoke at the company’s Future Decoded event in London at the start of November 2016.

“Microsoft wants to empower every organisation and every person on the planet to do more,” said Bishop. “Our approach is to use AI to empower people to achieve more. AI should augment human abilities.”

Read more about deep learning

For the last few years, Microsoft has been looking at using AI to make it easier for people to handle the workload modern digital communications have enabled. Delve was the company’s attempt at putting intelligence into threaded conversations. But with Microsoft Team, the company is beginning to add some of the ideas coming from its research labs, such as natural language processing and deep learning.

Bishop said Microsoft recently supplemented its datacentres with field programmable gate array (FPGA) chips. These devices can be wired and rewired programmatically, enabling them to perform simple calculations directly in the electronics of the circuit, which is far faster than a central processing unit (CPU) or graphics processing unit (GPU).

He said Microsoft had created the world’s first exascale AI supercomputer spread across 15 sites globally, based on FPGAs.

With the amount of power on tap from FPGAs, it is possible to translate the whole of the English-language Wikipedia site into Russian in one tenth of a second, said Bishop.

“The idea of using machine learning on large amounts of data is having a transformative effect across every industry and country”
Satya Nadella, Microsoft

In October 2016, the company applied deep learning and textural analysis to Leo Tolstoy’s War and Peace. The company’s AI researchers extracted key phrases from the 1,500-page book. They then used the Microsoft Text Analytics API to provide a way for a reader new to War and Peace with an interactive guide to the characters and plots.

At Future Decoded, the company demonstrated the power of its deep learning and AI research with live, simultaneous translation from French to English and Dutch. While the demo to 5,000 attendees did not work seamlessly, it did illustrate the confidence Microsoft has in its ability to apply deep learning to augment human communications.

“The idea of using machine learning on large amounts of data is having a transformative effect across every industry and country,” said Nadella, speaking at Microsoft’s annual Ignite developer conference earlier in 2016.

Thanks to the Azure cloud on which Office 365 runs, Microsoft Teams has the potential to capture business conversations and human interactions, applying deep learning to them as demonstrated with War and Peace.

From a business user’s perspective, Microsoft aims to enable Office 365 to present information in the best form, based on the work context. But with deep learning, Microsoft can potentially gain a huge insight into the inner workings of businesses.

Read more on Collaboration software and productivity software