Conversational UI isn’t a fad: a tale of 5 bots here to stay.
“Alexa, could you order more almonds please?”
Platforms that leverage conversational user interactions already fill niches apps do not. Each of these bots exemplifies interesting dimensions of the medium, both in what they do and in how they do it.
- CourtBot — lessening the burden on Atlanta’s courts
- Carebot — reporting, summarizing and delivering writers’ statistics
- x.ai’s Amy Ingram — AI scheduling assistant
- Alexa — Amazon’s in home bot
- Siri — iPhone’s built in assistant just became a platform
“Courtbot was built with the city of Atlanta in partnership with the Atlanta Committee for Progress to simplify the process of resolving a traffic citation. After receiving a citation, people are often unsure of what to do next. Should they should appear in court, when should they appear, how much will the fine cost, or how can they contend the citation? The default is often to show up at the courthouse and wait in line for hours. Courbot allows the public to find out more information and pay their citations immediately via a text message.”
About CourtBot Courtbot was built with the city of Atlanta in partnership with the Atlanta Committee for Progress to…www.codeforamerica.org
CourtBot is a brilliant project from Code for America that reduced the burden on Atlanta’s courts by allowing people to check the status of their court dates and pay fines over SMS. It was also internationalized into Spanish (Atlanta’s first internationalized service), a testament to the I/O simplicity of text based interfaces.
CourtBot exemplifies simple interfaces a single developer can build.
Watch Sam Hashemi present CourtBot here:
“Carebot cares about us so much it automatically reports out, summarizes and sends us our analytics.”
For years, journalists have railed against pageviews and uniques, twin metrics that have established supremacy over…www.poynter.org
Carebot, funded by the Knight Foundation and built by NPR, is a pragmatist. It goes to wherever busy journalists are and tells them about their reader statistics and other analytics, rather than expecting them to crunch data in complex metrics interfaces. Since it’s just outputting text via an API, any number of clients can be built on top of it (Slack, SMS, etc.)
This is a promising direction for BI more broadly. If we want people to understand and engage with data to make decisions, why not spoon feed it to them in a way they understand (text) in a medium they already work in (Slack)? This is a step forward from the status quo of asking people to regularly navigate on their own to an analytics dashboard that requires specialized knowledge to extract information from.
This is where Machine Learning comes in. We can’t write a program full of business rules that would understand those emails well enough to enable Amy to schedule your meetings. Scheduling a meeting is simply too complex for this kind of deterministic logic. On average, it takes eight emails, including the invite, to schedule a single meeting. Each of those meeting conversations contains many pieces of relevant information, with an intricate set of dependencies specific to that particular conversation. The vast scale and heterogeneous structure of all of that data makes it impossible to write enough rules to encapsulate, “What does an email about a meeting mean?”
x.ai’s Amy Ingram is an exemplar model of just how tricky it is to do natural language right, and what a monumental task getting a machine to flawlessly understand quickly dashed off messages is. x.ai employs humans to read, annotate and categorize messages; they double check the bots’ work to make sure scheduling happens right. They’ve got a team of data scientists and developers as well, and 10’s of millions in funding. If you expect to go beyond a simple menu of options for your bot, beware: the combinatorics are not in your favor.
Deep learning is a field within machine learning which uses algorithms that contain many layers of processing and…x.ai
We are fast moving from the app era to the era of the intelligent agent. Over the next half decade, we'll witness the…x.ai
I’m not bullish on the “smart home.” We’ve now seen the launch of a bevy of expensive products with marginal utility and shoddy security.
Rather than an insecure, expensive, complicated automatic coffee maker you control with an App, think coffee purchased with a simple statement when I realize it’s out. The action of making coffee with a French press will probably be the same in 100 years. It’s not the pleasing, tactile rhythms of running a home people want to ditch, it’s the operational logistics.
Automating home logistics is more possible than ever with Alexa, and this week the product catalog expanded enormously:
It's been possible to use Amazon's Echo and Alexa assistant to make basic purchases - reorders of existing products, a…www.theverge.com
As of writing, all products purchased via Alexa are eligible for free returns, to encourage users to explore the new medium.
One of the most tantalizing tech rumors in recent weeks has centered around The Information’s report that Apple is not…motherboard.vice.com
Finally, Apple is opening up Siri to developers as a platform. This is a step that parallels Amazon’s trajectory with Alexa. Now, developers from companies all over the world can build voice commands into Siri. “Siri, what’s my checking account balance?”
This is a huge boost to reaching users via natural language, and thus boosts the investment case for ‘bot helper’ technologies of all stripes, from platforms like Houndify to frameworks like wit.ai and hosting providers like BeepBoop. Expect the developer ecosystem to improve rapidly along with the platforms.
So forget wifi connected fridges and omniscient personal assistance and bots that tell stupid jokes. Instead, think: “Alexa, please order more laundry detergent.” That’s the power of conversational interfaces: a host of technologies that integrate seamlessly into our environment, don’t require us to ‘go’ to them but are ever present, and drastically reduce our time and energy to do repetitive logistical tasks — whether it’s checking in on the status of an article as a journalist, paying a parking ticket or ordering a refill on diapers.