Cloud

Google Cloud encourages more conversational AI with Bot-in-a-Box


Hear from CIOs, CTOs, and other C-level and senior execs on data and AI strategies at the Future of Work Summit this January 12, 2022. Learn more


Google’s business platform services just became a lot more conversational.

Google Cloud announced yesterday a new AI-powered service product, Bot-in-a-Box, which is a Google Cloud Platform (GCP) Business Messages feature designed to assist enterprises in initiating conversations with customers. GCP Business Messages is a conversational messaging service designed to enable organizations to connect with people to answer questions that come through Google Search, YouTube, Gmail, Google Maps, or their own business channels.

Bot-in-a-Box uses natural-language understanding and Google’s Dialogflow software to create chatbots that can understand and respond to customer questions without developers needing to write code. Using machine learning to understand a customer’s request, Bot-in-a-Box features Custom Intents, which finds the information a customer needs without human intervention.

Dialogflow CX is a virtual agent that handles concurrent conversations with a company’s users. The module is designed to understand the nuances of human language. Dialogflow translates end-user text or audio during a conversation into structured data that apps and services can understand. Bot-in-a-Box users can design and build a Dialogflow agent to handle the types of conversations required for their business. A Dialogflow agent, similar to a human call center agent, can be trained to handle expected conversation scenarios, and training does not need to be overly explicit, according to Google.

[ Related: How Zowie Diagnostics aims to refresh customer service chatbots with AI ]

Custom Intents is a Google product that aims to allow marketers to target people currently researching specific topics, products, and solutions on the web, using display or YouTube video ad campaigns.

While in the process of serving a customer, a Bot-in-a-Box also can display an existing customer FAQ document, whether it’s from a web page or an internal document, to keep the service simple and easy to follow.

Beta users of Bot-in-Box have included Walmart, Tango Technology, Levi’s, and Albertsons. They have utilized AI-powered Business Messages since its launch on all platforms in February, Google said, reporting that the automation of customer conversations has helped employees save time.

What’s new with this bot

Bot-in-a-Box represents the initial first-party integration between Dialogflow ES and Google’s Business Messages, Google Group product manager Jay Akkad said. The integration is designed to cut the time and resources required to create an automated experience for Business Messages. This makes it convenient and easy for any business to automatically respond to a portion of their customer support inquiries, Akkad said.

[ Related: Equipping AI with emotional intelligence can improve outcomes ]

No-code development is a key ingredient in Bot-in-a-Box for companies that deploy line-of-business employees to create bots. The first-party integration helps a non-expert connect a Dialogflow automated conversational agent to Business Messages through a series of UI forms. Once the connection is made, an existing FAQ can initialize the conversational agent. Dialogflow’s web-based console can also be used to extend the conversational agent to handle more complex questions and user journeys. The console makes it easy to create new intents that can match and respond to unseen user inquiries. All of this can be set up without writing any code.

Dialogflow’s fulfillment feature is available as a low-code option for creating dynamic responses based on backend business data. However, the intentional creation and matching can still be configured completely through an easy-to-use web-based console, Akkad said.

Helping customers in ‘moment of need’

Using AI-enabled Business Messages, businesses can connect with customers in their moment of need, in the places they’re looking for answers — such as Google Search, Google Maps, or any brand-owned channel, Akkad wrote in a blog.

“People turn to Google when they are searching for answers to their questions, looking to buy something, or trying to accomplish a particular task with one of our many tools. In fact, 68% of all online experiences begin with a search engine,” Akkad said. “We know how important it is for interactions with a brand to be personalized, helpful, and simple.”

How the AI components come together

In order to help technologists understand how Google is using AI to enable Bot-in-a-Box, here are some data points from Akkad. Naturally, GCP is where everything is housed and operated.

VentureBeat: What AI and ML tools are you using specifically?

Jay Akkad: Dialogflow is used to handle intent matching and configuration of responses. Additionally, Dialogflow’s knowledge connectors feature is used for FAQ upload and management.

VentureBeat: Are you using models and algorithms out of a box — for example, from DataRobot or other sources?

Akkad: We use Dialogflow’s built-in ML and NLP technology as well as their knowledge connectors feature.

VentureBeat: How are you labeling data for the ML and AI workflows?

Akkad: This is handled automatically by Dialogflow.

VentureBeat: Can you give us a ballpark estimate on how much data you are processing?

Akkad: This is dependent on the conversational agent that a business creates.

VentureBeat

VentureBeat’s mission is to be a digital town square for technical decision-makers to gain knowledge about transformative technology and transact.

Our site delivers essential information on data technologies and strategies to guide you as you lead your organizations. We invite you to become a member of our community, to access:

  • up-to-date information on the subjects of interest to you
  • our newsletters
  • gated thought-leader content and discounted access to our prized events, such as Transform 2021: Learn More
  • networking features, and more

Become a member



READ SOURCE

This website uses cookies. By continuing to use this site, you accept our use of cookies.