Microsoft will blend artificial intelligence (A.I.) with Excel, one of its longest-running products. Of course, it remains to be seen how merging relatively untested technology with a well-established platform will actually work in the real world. A.I. algorithms will highlight patterns within users’ datasets, streamlining the process of analyzing data. By alerting those users to outliers and other interesting data-bits, the theory goes, Excel can improve the larger business. Microsoft is extending A.I. to other parts of Office, as well. Word will have a new tool called Acronyms, which (according to the company’s official blog post) “helps people understand shorthand that is commonly used in their own workplaces by leveraging the Microsoft Graph to surface definitions of terms that have been previously defined across emails and documents.” Compared to other proposed A.I. uses such as autonomous driving, Acronyms doesn’t seem like a heavy lift; but there’s always the possibility of the system reading a jokey office email thread and accidentally alerting the CEO to an Urban Dictionary definition as opposed to the acronym for, say, Bounce Rate. Last but not least, Microsoft is integrating A.I. into Outlook that will automatically surface trips and deliveries, future meetings, and other items of note. Again, this isn’t a radical upgrade so much as Microsoft attempting to leverage a nascent technology to make an existing service a little more frictionless. Excel “Insights” will roll out in December, with other features due sometime in 2018. Whether or not you use Excel and other Office tools, Microsoft’s approach to A.I. in this context possibly heralds how many companies will integrate this sort of technology in the years ahead. Rather than focus on how A.I. can drive seismic change—again, as in the case of self-driving cars—companies may apply it in a more iterative way, making select features “smarter.” Many of these changes may end up so subtle that users barely notice, even if the new algorithms are quite sophisticated. For tech firms figuring out how to blend A.I. into their offerings, this is perhaps the best way forward: rather than expend tons of resources and time on so-called “revolutionary” change, iterative upgrades that make existing products “smarter” may ultimately prove more beneficial with regard to user retention and revenue generation. With vendors offering more A.I. tools, the possibilities for creation are only multiplying.