Susan Hendrich was talking with a friend who mentioned how hard a co-worker with dyslexia worked to ensure that meeting notes he was responsible for writing were error free.
Because he worried about making mistakes, he routinely brought meeting recordings home with him and labored over the summaries for hours at night.
That led Hendrich, a Microsoft group program manager and expert on natural language and AI for Microsoft Office, to wonder how her team might improve those stressful writing experiences for people whose brains process letters and words differently.
A new feature now being rolled out in Editor in Word can use more sophisticated AI to offer suggestions for rewriting full sentences rather than offering spelling or grammar fixes one at a time. In internal evaluations, it was nearly 15 percent more effective than previous approaches in catching mistakes commonly made by people who have dyslexia, Microsoft says.
That’s largely because the deep learning algorithms that can offer those rewrites were trained on large and diverse datasets, including documents written in the real world by people with dyslexia, rather than a narrow and finite set of linguistic rules.
“It was actually something that we did not expect,” Hendrich said. “But then we started doing some benchmarking and realized this can be a huge benefit to people.”
Microsoft engineers have steadily been incorporating more AI advances into its Microsoft 365 suite of products over the past several years. They now help over 200 million Office 365 users have more productive meetings, stay on top of their to-do lists, deliver more powerful presentations, preserve focus time and help people find ways to write more clearly.
Now, Microsoft is making its AI-powered writing assistance tools more widely available to enterprise and consumer customers around the world. With new features that begin rolling out today and will continue over the coming months, Microsoft Editor will give writers the option to use intelligent tools to craft more polished prose in documents, emails and posts across the web on sites such as LinkedIn, Gmail, Facebook, Twitter and more.
The new tools are available through Editor in Word, Editor in Outlook and a new Editor in the browser extension, which will allow users to catch mistakes and write more confidently when crafting social media posts or communicating elsewhere on the web.
Editor’s spelling and basic grammar checks will be available to everyone, and Microsoft 365 subscribers — including the new Microsoft 365 Personal and Family productivity subscriptions announced today — can opt to use more advanced AI features that offer intelligent suggestions for making writing more concise, clear, formal and more.
The releases expand Microsoft’s advanced AI-powered writing suggestions to Outlook.com, Outlook for the web and the new browser extension, currently available for Microsoft Edge and Google Chrome, in more than 20 languages. Basic spelling checks in the browser extension will be available in 89 languages.
Starting with Editor in Word and coming soon to other platforms, Microsoft 365 subscribers can also opt to see inclusive language suggestions — such as the word firefighter instead of fireman — that seek to eliminate biases based on gender, age, ability and more. In Word for the web in English, those users who want more feedback can now access whole sentence rewrite suggestions to improve fluency, conciseness and readability.
Other new AI-powered Word features, which will be expanded to other Editor offerings, can offer intelligent suggestions for spelling out acronyms, putting numbers in real-world perspective and even flagging potentially unoriginal language and allowing writers to properly cite reference material.
Susan Hendrich, an expert on natural language, leads engineers who infuse AI into Office products. Photo by Dan DeLong.
Massive scale and opportunity for impact
As Microsoft has incorporated more AI technology into its products, company executives say they’ve been keenly aware of their duty to help ensure that AI is used in ways that are responsible and thoughtful, that anticipate unintended consequences and that help meet real customer needs.
“If you think about the number of people who use Office in their daily lives, what’s unique about us is the massive scale and opportunity for impact,” said Sumit Chauhan, Microsoft corporate vice president for engineering in Office. “The burden is really on us to get this right.”
For instance, Editor in Word’s AI-powered rewrite suggestions tool has the potential to benefit many authors — from people learning English whose omissions of “a,” “an,” or “the” aren’t always caught by traditional language checkers to busy professionals who get hundreds of emails an hour and don’t have time to labor over each response.
At the same time, AI has the potential to reinforce harmful biases that exist in society, and in the data that algorithms learn from. To address this unintended consequence, Microsoft researchers have developed a block list of words that are at greater risk of causing offense, and Word doesn’t show rewrite suggestions with those phrases.
For those who choose to use it, Editor’s inclusive language critique offers suggestions to replace language that may perpetuate biases around age, ability, gender, sexual orientation, religion, ethnic or racial slurs, as well as outdated or sensitive geopolitical references.
The goal isn’t to correct all of society’s issues, Chauhan said, but to highlight potential blind spots and offer some general suggestions to consider.
If you think about the number of people who use Office in their daily lives, what’s unique about us is the massive scale and opportunity for impact. The burden is really on us to get this right.
“In our products, we want to be thoughtful about stereotypes in the world and to make sure our AI is not reinforcing or amplifying those,” Chauhan said.
Of course, those sensitivities vary widely by country and culture. That’s why Microsoft hires native speakers and linguistic experts in 20 languages to write the rules that guide grammar, clarity and conciseness checking — and to advise which inclusiveness critiques don’t have comparable solutions in that language or might be unwelcome in certain markets.
Considering users’ needs can be as simple as recognizing that people who are colorblind can’t distinguish between words written in red or blue and may also need dashed lines or squiggles to flag words or phrases.
Or it can be as complicated as figuring out how to handle gender-neutral pronouns in spelling and grammar critiques — something the team is experimenting with. The cultural shift to use pronouns such as “they/them” to refer to people whose gender identities are nonbinary is recent enough that there’s little data that can teach algorithms their proper usage.
For now, Editor in Word and Outlook will not flag the use of “they” in a singular context as incorrect — such as “they are taking their final exams tomorrow” rather than “she is taking her final exam tomorrow” — which Hendrich believes is a good first step toward including a wider spectrum of gender identities. But she also worries about the person who might use “they” by mistake in a cover letter and doesn’t land a job because Word’s grammar checker failed to flag it.
“We have a goal of empowering our customers to write confidently and feel like their writing reflects positively on them,” she said. “And we also have a goal to properly represent and celebrate people for their uniqueness. So, to be perfectly honest, these are the questions that I grapple with as a product creator. And I don’t always have the answer.”
Penny Collison, principal design research manager for Office, worked with researchers to develop best practices for AI-human interaction that guide designers as they develop M365 products. Photo by Dan DeLong.
Putting people at the center of AI design
Mira Lane, the lead of Microsoft’s Ethics & Society team, is charged with ensuring that the principles articulated at the highest levels of the company to guide the responsible use of AI find their way into the heads of researchers conducting user testing and the hands of engineers writing code. It starts with asking the right questions, she said.
Her team of philosophers, engineers, security experts, designers and trainers works closely with product teams to consider what data or models should be used, who might be directly or indirectly affected by a new technology, what kinds of people should be interviewed to identify unintentional harms and how those insights can be folded into product design.
“The thing that we’re trying to do is help people design technology in a really intentional way, so you really understand what the effects of the tech are and can look around the corner to how it might be used or misused,” Lane said.
For teams that incorporate AI into productivity tools, one of the most important principles is to keep people at the center of the process.
“We bring a lot of focus to making sure the experiences we’re delivering are actually valuable,” said Penny Collisson, principal design research manager for Office. “We have lots of conversations with customers where we never even mention AI. We’re talking about understanding the expressed or latent needs or pain points that people have and then we go back and try to think about how AI could fit in.”
Microsoft has developed 18 best practices that researchers and product designers use to guide their work. But a lot of that work involves listening to people with different levels of tech adoption, socioeconomic backgrounds, geography, physical abilities or attitudes about AI and privacy.
If you talk to people with learning disabilities, for instance, some have a fear of starting with a blank page. That insight helped guide improved dictation offerings in Word for the web, which makes it easier to create content with one’s voice and use speech-to-text to get thoughts down on paper.
Creating good user experiences with AI is more complicated than asking people for feedback on whether they prefer one type of control over another, or which interface is easier to navigate, said Jon Friedman, Microsoft corporate vice president for design and research.
“The kicker and power of AI is that everyone’s experience is unique. Before, we were designing for the mean because solutions were closer to one size fits all. And now we are designing each thing to be a special size to fit each individual,” Friedman said.
“So making sure we’re talking to a much broader set of people and hearing everyone’s voice is really important to give people what they truly need,” he said.
In one example, Microsoft designers and engineers who were interested in building a better screen reader for people who are blind or with low vision built a relationship with the Washington State School for the Blind and began interviewing and observing how those students consume information and approach tasks for the day. That work led to Play My Emails in Outlook mobile, which turned out to also be useful for anyone who wants a jump on their day but can’t safely look at a screen while commuting or cooking breakfast for kids.
Through interviews and equipment that simulated the experience of having macular degeneration, the design team began to understand the massive cognitive load that’s required to listen for pertinent information among a sea of extraneous details like dates and time stamps and even punctuation marks that screen readers include as they scan from left to right.
“It was like listening for a needle in a haystack, and the fatigue level was really high,” Friedman said.
So the team used AI to offer the most important information upfront and in a much more conversational way. Having Cortana, Microsoft 365’s personal productivity assistant, tell you that someone sent you an email in the past hour about scheduling a meeting this afternoon is more useful than knowing the precise time stamp, Friedman said.
Play My Emails also provides summary information like how many unread emails are in your inbox and how long it would take to listen to them. That helps people decide if they have enough commute time or brain space while they’re rushing to get out of the house to focus on the task.
“We started on this path because we thought inclusive design was an important philosophy that we needed to start living and breathing in product,” Friedman said. “But the team quickly realized that there’s a lot of instances where people are situationally blind or looking at screens when it’s not safe, and that’s when they realized this is something that could be useful for people in a lot of different contexts.”
Jon Friedman, Microsoft corporate vice president for design and research, says the power of AI to deliver unique experiences to each individual requires a more thoughtful approach to user testing. Photo by Dan DeLong.
Keeping users in control
Microsoft also wanted users to have control over its new AI-powered tools. Some of the new Editor features, like inclusive language critiques, aren’t automatically turned on but allow people to opt into using them.
Researchers also heard loud and clear from some users whose unique writing style is an important part of their brand or professional success. They didn’t want AI algorithms rewriting their prose. That’s, in part, why the Editor in Word rewrite suggestion tool includes three options that can serve as inspiration or jumping-off points for an author to build on.
“In some ways this is just exposing the idea that there are different ways to say the same thing,” Friedman said. “We believe that’s going to help with both the concept of teaching people and letting them have their own voice.”
Designers of the Editor tools regularly talk to teachers about how their students prefer to receive editing feedback and how the tools can help build students’ skills rather than simply doing tasks for them.
“Teachers don’t just want us to fix their students’ work,” Hendrich said. “They want them to be able to learn in the process. That’s why Word offers explanations, to help people write more confidently while also providing these teachable moments.”
That was one of the drivers behind the new similarity checker function in Word for the web, which can help flag unoriginal content that may have made its way into a document from the abundant reference material available online. The similarity checker also helps teach people how to properly cite others’ words or ideas, which is an important skill for students to learn and one for which teachers have said they would welcome extra help.
As Microsoft looks to the future, teams are exploring other new features made possible by advances in natural language processing, computer vision and other types of AI. That includes Presenter Coach features in PowerPoint that can tell people how often they slouch or make eye contact during a presentation. Tools in Office that help people accomplish more of their routine tasks on the go through voice interaction can free up time for more focused work.
“When you’re using our products, you’re trying to get your job done right,” Chauhan said. “Our job is to amplify that human ingenuity. So whether you’re writing a document or designing a beautiful presentation or analyzing a spreadsheet, we’re always thinking about how can we help you in that task. That’s really been the arc of AI in Office.”
Top image: Teams led by Sumit Chauhan, Microsoft corporate vice president for engineering in Office (left), and Mira Lane, who heads Microsoft’s Ethics & Society team (right), collaborate to ensure that AI-powered productivity tools are designed with the company’s values in mind. Photo by Dan DeLong.
Related:
- Introducing Microsoft Editor – Bring out your best writer wherever you write
- Introducing the new Microsoft 365 Personal and Family Subscriptions
- Extending the power of Azure AI to Microsoft 365 users
- How AI is making people’s workday more productive
- AI and Cortana in Microsoft 365 put people at the center
Jennifer Langston writes about Microsoft research and innovation. Follow her on Twitter.