How to get started in Conversational AI

Picture of Kane Simms
Kane Simms

This question comes up all the time, and that’s great! So many people are discovering conversation design, and so many of them come from different backgrounds.

We want to give you some helpers to get started.

Before we do that, it’s worth pointing out a few things.

  • Conversational AI has the potential to help everyone, but in order to achieve that we need people from all walks of life
  • Most people have an unrealistic view of what is currently possible (probably because the internet is awash with hype about AI)
  • This field is constantly evolving – you’ll be fine if you’re curious and can adapt to change

The good news is that while the possibilities for conversations between humans and machines are evolving, many of the underlying principles stay the same! So, we recommend you check out the following to get started.

Understand how language works

Just because you can communicate doesn’t mean that you can analyse and design conversations. It’s important to learn what you can about how conversations work. The following are some great resources

Get a grounding in AI

Not every conversational experience is based on AI, but it’s almost always involved in some capacity. Over time it’s very likely that AI (or machine learning to be more precise) will become more intertwined with our work.

Read about conversational AI

Books:

Online:

Get training

There’s many courses available, and different options depending on your experience, aims, budget and so on. Most are taught in English.

Start making and learning

It’s great to read and learn from others, but that knowledge is going to get supercharged when you start to apply it yourself.

  • Sign up to ChatGPT and begin prompting – all you need to do is communicate with it, and observe how altering your inputs affects the outputs you get. While your experience appears to be communication between two beings, ChatGPT is just a statistical model, trained on things people have said and written in the past. How can you work with the model to get a result? What concrete changes can you make that improve the truthfulness and accuracy of the output?
  • Sign up to Voiceflow and start making bots – here you can practise common conversation design techniques. Think about the flows that are needed for your bot – what will people need to know, and how will they interact with it? How will it feel? How can you improve it? VoiceFlow is a popular tool that is constantly improved. They’ve recently added generative AI features too.
  • You should write up your research and creations to add to your online portfolio as well! Describe your process and your learnings (your final result is less important to the reader than how you got there).

Attend industry events

There’s an event for everyone – from evening networking meet-ups to multi-day conferences with workshops and unique content. Events offer a plethora of opportunities to learn, network and keep up with the latest best practices. Most of the industry events can be attended virtually, so you can participate from the comfort of your home or office.

Some of the events in the Conversational AI space include:

Final words

Don’t assume that anyone has all the answers. We’re all learning here, and things seem to change rapidly. You might come to conversational AI with some ideas that push the whole industry forwards.

Good luck!

Share

Weekly newsletter

The latest in AI-powered customer experience. Make better strategic decisions with the help of our weekly newsletter.

Kane Simms

A strategic AI advisor who, for the past decade, has helped business leaders and product owners transform customer experience using conversational and generative AI.

Share

✓   Link copied

Weekly newsletter

The latest in AI-powered customer experience. Make better strategic decisions with the help of our weekly newsletter.

Top articles and podcasts

Related content

Cobus Greyling: why intents, graphs and conversation design have come full circle
Cobus Greyling on where enterprise AI is really heading: data, security, forward-deployed engineers and conversation design.
From answers to actions: building better government AI
Pete Kowalczyk on knowledge graphs, goal detection and moving AI from answering questions to taking action.
What’s really breaking your AI agents?
Discover where conversational AI projects actually break: contradictory source data, manual process and weak testing.
How Simplyhealth scaled voice AI to resolve 30% of calls
Paul Atkins, Head of Business Optimisation at Simplyhealth, breaks down how they scaled voice AI to resolve almost a third of calls, and what it took to get there.