Quick. Casual. Human. Chat differs from other forms of communication. Because of these three attributes, chat seems to be reemerging as a potentially disruptive user interface for both consumers and business users.

The typical teenages emits more than 3000 text messages per month, not to mention messages on other networks. Most retailers provide customer support via chat. Hundreds of millions of people have tweeted. The combination of these three forces expose nearly everyone on the Internet to brief, useful messages every hour.

It’s not that chat is new. Internet relay chat dates back to the earliest days of the Internet. But, the exciting advances promised by new startups extends much further than message exchanges between two people. These chats have consequences. Send a text message to book a flight. Post a note on Slack to provide peer feedback at work or file an expense. Wire $15 to a friend to cover last night’s beers.

In many of these use cases, chats replace forms, the list of empty text boxes we must complete in order to accomplish almost anything on the Internet or mobile phone. In fact, most SaaS software today has two principal components: an input form and some analytics that aggregate those data across time. And who doesn’t sigh every time they login to Salesforce to enter a new customer?

Imagine a user interface that asked about your day. Which customer did you meet today? How much budget do they have? What is their timeframe for buying? Questions, a fundamentally human construct, replace the lifeless, empty cells of a database recreated on a webpage. A question engages users the way another person would, even if the person asking the questions isn’t a person, but a robot.

Machine learning and large data sets enable this advance in user interface. Given enough data and sufficient computational power, computer scientists can code bots that, at least for some fraction of the time, can replace humans asking questions.

Intelligent, question-driven chat will become an important user interface for many different use cases, with a few characteristics in common. Frequent use, because chat requires learning a new user interface, and behavior change demands repetition. Brief and relatively simple interactions. Users won’t pen Powerpoint presentations in 140 character increments. Multi-platform applications that work just as well on the phone as on the laptop and tablet. Applications that benefit from a more casual interaction.

Chat software startups will face several key challenges when coming to market. First, SaaS applications rebuilt as chat-only user interfaces must provide more than the current competing offerings to succeed. Second, if integrating with existing messaging platforms, these chat businesses will face platform risk. The platform company providing the messaging infrastructure may decide to pursue a particular use case themselves. Third, generating meaningul revenue per user with a smaller price anchor. How valuable is a user experience advancement to most software buyers? Unlike the Salesforce ecosystem in which buyers compare sales productivity tools to the $150 of a Salesforce seat, the prices in the messaging ecosystem are much smaller - just a few dollar per month for Slack or Office 365 (which bundles Yammer).

None of these are insurmoutable obstacles, just another set of considerations. There’s a wave of innovation looming with chat user interfaces powered by artificial intelligence (with human fallback), and we’re just at the start of it.