The vision behind Activechat is to make the most advanced AI technology easily available for customer care applications even if your team has zero technical background. This vision empowers our natural-language integrations, sentiment detection, and other advanced features packed into our visual chatbot builder. But there’s more to it, and we’re happy to introduce our recent integration with OpenAI’s beta of their GPT-3 model – GPT-3 for live chat.
GPT-3 is a mega machine learning model, created by OpenAI, and it can write it’s own op-eds, poems, articles, and even working code. As a result of its humongous size (over 175 billion parameters), GPT-3 can do what no other model can do (well): perform specific tasks without any special tuning. You can ask GPT-3 to be a translator, a programmer, a poet, or a famous author, and it can do it with its user (you) providing fewer than 10 training examples. Damn.
Dale Markowitz
Activechat has partnered with OpenAI team to build something that looks like real magic. Once you describe the context of your business in a plain natural language and provide a couple of example questions and answers that your human live chat agents may encounter in their day-to-day conversations with customers, we can use the power of GPT-3 to provide instant hints to your agents to help them answer almost any question that a customer may have.
Let’s look at some real-life examples of this amazing tech.
Bike shop – default settings
Imagine that you’re the owner of a bike shop selling hundreds of bikes from various manufacturers. Your website has a lot of traffic and you introduce live chat to help your customers make their choice and grow your revenue. After a week of accepting incoming chat requests, you suddenly realize that training your agents so that they can really answer this crazy amount of questions from customers-to-be takes tons of time and is quite costly. You have your knowledge base, but most of your agents feel lost when it comes to answering specific questions, and looking for the correct answer often takes quite a lot of time.
Luckily enough, your live chat solution is powered by Activechat, so you can go to “Settings – Integrations – OpenAI” and describe your business case in plain natural language:
This is a conversation between a website visitor and a smart virtual assistant. The conversations are happening across various pages of the website that is selling bicycles, and virtual assistant can answer all kinds of questions about various bike brands and models and help choose the right one according to description made by the visitor.
That’s all you need to seed the OpenAI’s GPT-3 for live chat
Next, you provide answers to three questions that your customers ask most often:
Q: What’s the difference between a carbon and regular bike?
A: The primary difference between carbon and aluminum comes down to weight and ride quality. Carbon frames are usually a bit lighter than aluminum — up to a pound for mountain frames and up to a half-pound on road frames.
Q: How much does a good bike cost?
A: Road bikes range between $350 and $700, Mountain bikes around $1000, Single-speed bike – $400, Beach cruiser $200-300, Recumbent bike $1000-2000, and Kids’ bike (3-8 yrs) will be $140-200.
Q: How do I service my bike?
A: You should regularly service your bike to ensure it is running efficiently and that there are no worn or damaged components. The more you ride the bike the more frequent you should be servicing it, especially after riding in dirt, sand, mud and in the rain.
Guess what? Your training is complete!
Now your agents can pull the “Show AI hints” tab in their live chat interface, and on every new message from a website visitor, there will be 2-3 ready-made answers suggested by the GPT-3 engine.
Clicking any of these hints will copy the answer to the message window, and your agents can send it immediately or edit it, adding specific links to your product pages or other resources. Based on our research, this can reduce the time needed to find and type the answer by 70-80%!
If you’re not satisfied with what you see, just hit the “Refresh” icon and a new set of hints will appear. These hints are powered by all knowledge that the Internet has (GPT-3 model is trained on Common Crawl data, Wikipedia, and multiple other data sources, including a huge amount of books). This makes suggested answers sound human-like and in most cases, they will be relevant, useful, and valuable to your customers.
AI-powered live chat for mobile phones outlet
Another use case from one of our customers – an online shop selling mobile phones. Again, the problem is quite similar to the bike shop above – customers repeatedly ask questions that require a solid chunk of knowledge from customer service agents. Novice agents keep forwarding these questions to more experienced team members, and instead of taking care of really complex queries, these team members keep wasting their time responding to the same questions again and again.
After they introduced GPT-3 for live chat, their seeding settings would look like this:
And here is an example conversation that was made completely by sending GPT-3 answer suggestions, with zero time required from the human agent:
How to use GPT-3 for live chat?
Currently, OpenAI’s GPT-3 is in a public beta. Activechat partnered with OpenAI team to include these amazing features into our live chat platform, and they are available for all our users on Team and Company plans. Here’s how to jump-start your AI-enabled live chat:
- Go to “Settings – Integrations – OpenAI” in any of your bots and type your business description in plain natural language.
- Add three questions that are most relevant to conversations happening on your website.
- Ask your live chat agents to pull the “Show AI hints” tab below the message editing window.
Yep, it’s that simple! We can’t wait to hear about your use cases, so please feel free to share your stories of GPT-3 making life of your human live chat agents easier.