How Bots, Messaging & AI Are Evolving

What 7 insider experts like Chris Messina, Shane Mac, Esther Crawford, Matt Schlicht & more think about the future of bots, messaging & AI.

Published in

Chatbots Magazine

13 min readOct 11, 2016

Bots, Messaging & AI have been a super hot topic within the startup and tech world for the past few months — and it’s all really only getting started as messaging and chat platforms are expected to continue to grow quickly.

“Most people don’t know this, but messaging apps are growing faster than social networks”

— Matt Schlicht, Founder & Editor, Chatbots Magazine

Projected Growth of Mobile Messaging Users. Source: http://www.statista.com/statistics/483255/number-of-mobile-messaging-users-worldwide/

From Snapchat, to Kik, to Facebook Messenger to Telegram, to Slack and more — building a platform on which to build and scale bots is the modern browser wars of the late 1990s.

Tyler Copeland & I have been interviewing hundreds of awesome people who are out there making things happen for our podcast called Hack To Start (shameless plug & context for this post!).

Naturally, we’ve come across some of the leading entrepreneurs and investors in the bots and messaging space. They are:

Here’s a a quick recap of their thoughts and insights on the subject of bots, messaging & AI.

Ben Brown

Co-founder & CEO, Howdy.ai

Howdy is the company behind bot building tools like Botkit and was one of the first to start building Slack-bots with one of their own that goes by the same name.

Here are some the best insights from Ben on the subject of bots, messaging & AI:

[8:25] There’s a bunch of disparate technologies that are all maturing right at the same time. They ranges from things like these messaging platforms—like Slack, Kik, Facebook Messenger — to things like augmented reality and sensors, and internet of things.
[8:53] All of these things have been coming for a while and they’re now kind of finally arriving and opening up all these new opportunities to connect things and way to interact with things.
[9:30] Chatrooms, IRC, chat bots and things like that have always been part of the core internet experience for me. I’ve always loved that kind of stuff.
[10:22] We started building Slack applications because we thought it was cool and we thought it was fun. Then our clients started asking us to build Slack applications because they through it was cool and their teams wanted to use Slack more.
[11:27] It’s exciting because there’s a lot of interesting technology and techniques already out there from years and years and years of people doing stuff — none of which have ever really been applied to business software.
[12:12] That combined with the fact that—here’s this nascent platform which seems filled with enormous potential — and there’s like zero apps. There’s going to be an app store full of apps. We all get to be among the first right now.
[16:46] The basic tools for doing this needed to be available. Nobody is going to buy a bot if you can’t look at a store full of bots.
[19:00] If bots are going to be successful, you have to figure out all the little, low-level, weird little stuff that has to be build into those things. Like people say thanks to bots. And if the bot doesn’t understand how to process the “thank you”, it responds with an error message.
[20:07] We don’t even know — barely — about what people are gonna want to do with bots.
[22:58] As things continue to get connect to the internet, it’s hard to imagine even, how it’s supposed to talk to each other or interact with one another.
[23:23] Language is a really easy way for us to actually access this stuff and we barely use it in this context.

Matthew Hartman

Director of Seed Investments, Betaworks

Betaworks is one of the most innovative and creative investment groups out there. They literally pioneered the bot accelerator with the launch of Bot Camp.

Here are some the best insights from Matt on the subject of bots, messaging & AI:

[19:54] The idea around investments We focus on consumer tech, but particularly on user generated content in ecosystems around what we call internally “internet native media” — so things that can’t exist without the internet.
[25:48] Conversational interfaces — that can be anything from a text based bot to an Amazon Alexa app. I think we’re in the very earliest stages of that ecosystem developing and the new products that are gonna be big.
[27:11] It’s obvious to say that I’m going to automate a process, it’s less obvious to say that I’m going to use this bot to actually connect a whole bunch of people and do something efficiently.
[27:50] We are very, very early in, what I view as a kind of a fundamental shift away from Graphical User Interfaces into a different type of conversational interface.
[28:15] There are really great user experiences that can be fully audio. You then don’t even need to look at your phone anymore for certain things.
[28:24] The gap that we have right now is our ability as programmers to understand what people are saying — not only Natural Language Processing but also understand intent.
[28:33] We have a bunch of conversational interfaces that are pretending to be AI but they’re really not. And as the AI gets better and as the conversational interfaces become more ubiquitous, I think those two things combine to create really interesting bots and services that can just exist wherever you happen to be — and your kinda profile is almost in the cloud — and I think that’s a really exciting place.
[30:25] It’s as much about the network of people that are coming together on a platform, as it is about the technology itself.

Shane Mac

Co-founder & CEO, Assist

Assist isn’t actually an “app” — rather it works on top of existing platforms like Facebook Messenger and enables other apps like Uber to work inside messaging platforms.

Here are some the best insights from Shane on the subject of bots, messaging & AI:

[16:17] How do we turn every single service into a chat conversation? Where you can do it with text and you can do it without having to download an entire app or design this new UI…
[16:27] Businesses aren’t going to be able to continue to afford to build apps. Apps had this promise, but they never became easy enough for mass distribution like the web did.
[16:48] If we looked at Messaging as the browser, we’ve never seen a browser with this many people staring at it everyday.
[18:12] We made a decision last July to kill the human side — go all bots and automation and AI. Kill our own app. And just build across every messaging platform.
[19:00] Right now it’s completely insane — mostly because all of these platforms are changing so fast.
[19:30] It’s literally like the browser wars of 1998 — but worse. And so, as I say that, I think that’s the opportunity.
[19:40] I think every business in the world will have a bot interface that brings them to Messenger.
[19:44] If 80–90% of people’s time is spent within 5 apps — which are all messenger apps — every business is going to want to play there.
[19:54] I think there is a space between web and apps that is much bigger. And it will actually be the biggest space in the world in the next 5 years.
[21:00] Platforms — they really want to bring businesses to them — so Facebook really wants businesses on Messanger, same with SMS, and What’s App and Slack, and so they’re pretty open.

Esther Crawford

CEO, Olabot

Olabot is a personal chat bot platform used by celebrities and influencers like: Redfoo and Chris Messina alike. They are pushing the boundaries of personal bots and learning a ton about how people perceive and interact with them.

Here are some the best insights from Esther on the subject of bots, messaging & AI:

[23:19] As these messaging platforms have started to open up with Slack, and Kik, and Facebook, it’s just felt like this inevitability now that we will see companies, and products, and services, and mini apps live on top of these messaging platforms. They’ve become a new kind of operating system.
[31:42] He’s got this huge, massive social following — but he still struggles with how does he communicates with his audience in a more personalized way? And so, what happens when he not able to scale as a human — could potentially a bot do this for him?
[33:07] It’s such a fascinating thing to watch, and it’s so new. I mean this is the most nascent space right now with bots and AI. And watching this conversational paradigm unfold—I try literally a dozen or more bots a week and the rate at which they improve…
[34:00] The opportunity for the next 5–10 years of technology companies to be really build on top of this stack that’s being developed right now with AI, NLP, the piping that’s going to exist in this conversational format is going to spawn the next big wave of tech companies.
[35:20] It’s such a new experience and people don’t expect it. Like if a celebrity or somebody with a large audience posts a picture to Instagram or posts a Tweet, there’s not really an expectation that they’re going to send you a direct message afterwards […] and the bot paradigm sort of enables that.
[35:51] We’re trying to navigate that — like how do we let you know when it’s the person and when it’s not…
[36:13] It’s super interesting. I have a lot more questions then answers.
[36:24] Every bot that we release out into the world we just monitor really closely to see what works and doesn’t work.

Matt Schlicht

Founder & Editor, Chatbots Magazine

Chatbots Magazine is the #1 place to learn about Chatbots.

Matt Schlicht has a deep personal history with products and the internet so it’s no surprise that he was one of the first that saw the major shift towards messaging and decided to take action.

Here are some the best insights from Matt on the subject of bots, messaging & AI:

[21:24] Messaging platforms are actually growing faster then social networks. They’ve actually grown so much that at this point, they’re larger then social networks.
[21:43] This is crazy. This is a huge shift in what people are doing everyday.
[21:57] Every business in the world is going to have to come up with a strategy on how to interact with people on messaging platforms.
[22:05] The only scalable way for a business to interact with their audience or their customers on messaging platforms is with bots.
[25:57] I’ve done everything I possibly can to create a centralized area for bot developers, bot designers, bot investors, bot companies, and people who are interested in bots — where they can all communicate and share their knowledge and push the industry forward.
[34:45] What are the best practices for bots? How should bots be made?What are the best platforms for bots? How should you present your bot? How do you market a bot? There’s no standards in the industry right now.
[34:58] Bots are can incredible piece of technology if used right, but there’s a huge chance that they can be used wrong. […] There’s kinda this sense of responsibility that I have and that I’m trying to instill in the community.

Jeremy Goldberg

Product Designer, Facebook Messenger

Facebook acquired a little group messaging app called Beluga in 2011. It would go on to become Messenger.

With over 1 billion people on Messenger, Facebook is deeply interested in opening the platform to both developers and businesses alike, in order to create deeper relationships with their customers and audiences via chat.

Here are some the best insights from Jeremy on the subject of bots, messaging & AI:

[15:30] We’ve started to build messaging ecosystems that are incredibly expressive and rich. And we do that with people that we care about, with our friends, with our family. And we see this opportunity to bring businesses into Messenger and to create really meaningful relationships with them.
[16:57] There’s a lot of understanding to do. There’s a lot of understanding about how people already use the product, and how we can leverage certain situations or certain contexts to make those situations better with businesses or services.
[19:06] All these things we found were really important in how people perceived it — how they talked to it.
[20:22] There are a lot of high friction moments in our day — that doesn’t need to be. Through conversation and through simpler, more natural interactions, I think we can start to solve a lot of those problems. We can do it all within the context of messaging, where a lot of people spend a lot of their time.
[23:14] And on the other hand you have messaging ecosystems with are very conversational in nature and they’re very voice centric. We speak to each other in these platforms. And so, it’s only natural, I think that bots and a lot of these conversational paradigms start in a more voice centric way.
[23:40] I think it’s very important to understand that spectrum and where you would plot current software experiences.

Chris Messina

Developer Experience Lead, Uber

Chris has been a champion for the open web since the start of his career. He’s worked for companies big and small — and even launched a few of his own. He’s currently at Uber as a Developer Experience Lead — which was among the first companies to launch a “bot” feature in Facebook Messenger.

He invented the hashtag and successfully predicted the rise of ‘conversational commerce’ a term he coined to describe the way that conversational interfaces (like chat or voice), would lead to the framework for commerce or other services to be contextually injected in front of the user.

Here are some the best insights from Chris on the subject of bots, messaging & AI:

[24:44] It’s so much cheaper to prototype and build an app in that context [messaging] then it is to build a full fledged iOS or Android app.
[25:16] You’re starting to see how the conversational context, becomes this vehicle for delivering services.
[26:27] And instead of creating a GUI — you know a conventional pixel based interface layer — you can just talk to it. You can just express what you want.
[26:45] We’re trying to make software easier to use for people and conversation happens to be a very accessible, easy way to do that.
[27:27] App fatigue is real. Like, how many more apps do I need on my phone?
[27:55] You’ve got a really, really, really rich and interesting moment in the context of human computing that I think is really interesting.
[28:28] Most computing experiences have been designed for the single player experience. […] The history of social computing is relatively young, and so messaging by it’s very nature is communal and group based — much more easily then apps.
[29:17] In the messaging context, because you have multiple parties that are participating, now you can bring in one, two, or three or multiple bots.
[29:33] Now you start to see interactions occur in this shared context. And what this means is a couple things. One, it means that people will learn a lot more quickly. […] You can see how other people interact with a bot and then emulate that behaviour.
[32:12] Bot-to-bot sharing of information is another dimension that becomes incredibly interesting and opportunistic in a way that, currently app-to-app connections require app developers to think really hard about which apps you’re going to connect to and it then therefore limits the number of possible permutations of APIs that they might connect to.
[38:15] We’re now entering into another moment, where the messaging context is going to become a new publishing paradigm.
[39:14] It feels more accessible and like I’m talking to a real person, and so when I ask questions about what’s going on, I actually get a human response. So I think that’s really interesting, because it demonstrates why this platform is different and interesting and valuable relative to all the platforms that came before.
[39:39] In the case of messaging, you have an intimate, one-to-one interaction with the author.
[42:09] They’re really playing with character design, with interaction, with conversation design — there’s sort of a new emerging field called CUI; like conversation user interface — and it’s really interesting to see how they’re pushing the edge of these interactions. Where it almost feels like you’re talking to a real person.

Feature CBM: Designing a Chatbot’s Personality

[44:14] We’re gonna move into an era where our software and technology needs to become a lot more empathetic and to show, I don’t know more concern and caring for people. We’re sort of in this weird world where our technology is so cold.

Thanks for making it this far!

If you have any feedback, thoughts, ideas for future interviews with both bot entrepreneurs, influencers & investors you’d like to see—or just want to share what you’re working on, please comment below!

If you like what you read, please share and 👏— as a writer & podcaster it would mean the world!