![](https://d15shllkswkct0.cloudfront.net/wp-content/blogs.dir/1/files/2023/04/ai-robot-using-computer-chat-with-customer-scaled-1.jpg)
![](https://d15shllkswkct0.cloudfront.net/wp-content/blogs.dir/1/files/2023/04/ai-robot-using-computer-chat-with-customer-scaled-1.jpg)
Mobile messaging and conversation platform provider Sendbird Inc. today announced the availability of a new, ChatGPT-powered low-code chatbot API for developers.
It means that thousands of mobile apps that rely on Sendbird can now incorporate OpenAI’s humanlike conversational capabilities into their customer service chatbots. Sendbird is the creator of a popular application programming interface that enables mobile apps such as Reddit, Delivery Hero and Hinge to embed real-time chat, voice, video and notifications capabilities into their apps. It’s also one of the most popular tools for apps to integrate chatbots for customer service.
However, these chatbots are somewhat limited as they’re based on pre-programmed scripts. They can provide a series of standard responses for commonly asked questions but struggle to deal with more complex issues. That’s where ChatGPT could make a difference.
OpenAI LP’s ChatGPT has taken the internet by storm for its incredible conversational skills. Trained on millions of web pages available on the public internet, it can provide a humanlike response to almost any kind of question. Moreover, OpenAI enables companies to boost its capabilities by training ChatGPT on internal knowledge bases, so it can be applied to more specific customer service use cases.
Sendbird said its API can now act as middleware, giving app developers a simple way to access ChatGPT and use it as the foundation of their own chatbots. Businesses such as online marketplaces, healthcare providers, financial institutions and educational providers can create low-code ChatGPT-style chatbots trained on their own knowledge bases and data to enhance customer service.
It enables a more realistic and personalized chat experience, with the chatbot able to ingest information from internal customer relationship management software and prior conversations to provide more relevant responses to user queries. Sendbird said dozens of applications become possible.
For instance, healthcare providers can create a chatbot to handle doctors’ appointment scheduling while considering each one’s availability. Meanwhile, a financial services provider can create a customer service both to respond to a customer’s queries on their recent spending. Or an online marketplace can create a support agent with knowledge from previous conversations to deliver a more profound sense of customer care.
Sendbird co-founder and Chief Executive John Kim said the company’s mission is to help online businesses build deeper connections with their users and get things done. So it makes perfect sense to tap into ChatGPT’s capabilities and deliver more personalized experiences.
“The test cases that we’ve seen have been pretty amazing,” Kim said. “And like all Sendbird offerings, the API integration is remarkably fast and easy so that developers can integrate the chatbot into their apps right away. As such, we can lower the barrier to entry for customers looking to leverage ChatGPT.”
Customers can connect their existing chatbots to ChatGPT through the Sendbird dashboard, using their OpenAI API key. From the dashboard, developers will be able to develop effective prompts and set ChatGPT parameters all in one place. They can add any existing chatbot to the service, just as they would add new users.
Sendbird said the ChatGPT integration paves the way for it to support additional large language models in future, hinting that many more generative AI-powered use cases are on the way.
Kim appeared on SiliconANGLE Media’s mobile livestreaming studio theCUBE last June at the AWS Startup Showcase “MarTech Emerging Cloud-Scale Customer Experiences” event, where he spoke at length about how Sendbird delivers its in-app chat and messaging capabilities:
THANK YOU