Conversation design is about making interactions with AI assistants flow naturally and directing those interactions toward successful outcomes. It fuses technology, psychology, and linguistics to render our experiences with chatbots and voice assistants more human-centric. As you may have guessed, conversation design is experiencing a "gold rush moment" right now, as evidenced by the popularity of AI assistants like Google Assistant, Amazon Alexa, and Apple's Siri.
Chatbots and voice assistants, in themselves, are not exactly new. They've been around for a while now. What's new is their ability to interact so naturally with human beings - even when the prompts are less than clear.
In the early days, all chatbots could do was spew out canned answers to a subset of questions deemed the most common. If you strayed from the script ever so slightly, you were left with less than relevant answers to your query, or, in a customer service scenario, you could be transferred to a human being. But recent advancements in generative AI and natural language processing have changed the game. Chatbots and voice assistants today are able to understand and process a wider variety of input. That allows them to partake in human-like conversations that provide real value to human users while enabling humans to engage with them much more naturally.
But while technological advancements within AI have made chatbots and voice assistants "smarter," advancements in tech are only part of the equation. The tech is the canvas that supports the interactions. The other key ingredient is conversation design.
It's critical to understand that artificial intelligence and natural intelligence are two completely different things - with two completely different brains. A chatbot's or voice assistant's "brain" requires structured data sets (en masse) to work properly. A human brain also needs data that is structured in some form or another, but beyond that, it also critically needs things like empathy and understanding. In other words, a human brain necessarily emotes within the context of a conversation. An artificial brain does not - it simply predicts the most likely words to come after a prompt. So to get a successful interaction between a virtual assistant and a human being, we must provide both brains with what they need to work correctly.
That's the crux of conversation design.
Conversation design is so challenging because it sits at the nexus of the technical and the creative. So it's critical to define a clear workflow that covers both ends of the spectrum. The conversation designer needs to understand the inner workings of artificial and human brains and create conversation flows that provide each brain with what they need to tip the balance toward a successful conversational outcome.
And that's not a simple task.
But you can learn more at Modev's upcoming VOICE & AI conference, September 5th through the 7th, at the Washington Hilton in Washington, DC. We have some of the top talent in the field of AI ready to share their knowledge, experience, and insights into the subtleties of excellent conversation design.
Here are some highlights from a few we're sure you won't want to miss.
From Words to Impact: Best Practices for Prompt Engineering
Emily Banzhaf, Content Designer at WillowTree, and Marlinda Galapon, Principal Conversation Designer at ZenDesk, will discuss the process of optimizing instructions provided to AI models for successful outcomes - that's prompt engineering.
Together, Emily and Marlinda will take attendees on a bit of a journey, walking them through best practices for prompt engineering. And they won't simply be laying out the theory. They'll be sharing the stage with a few custom-built bots to provide concrete and practical examples of just how prompt engineering works, how to use prompt engineering for successful interactions, and how to avoid some of the potential pitfalls.
All the key components of successful prompt engineering (personas, testing, documentation, etc.) will be discussed, providing attendees with everything they need to optimize their conversation prompts and create the best experiences for their brands.
Voice First Multimodal Design: Do's and Don'ts
With over 20 years of industry and research experience specializing in designing, developing, and deploying natural speech and multimodal interfaces, it's probably fair to say Lisa Falkson knows a thing or two about conversation design. She's currently Senior Conversation Designer at Amazon, leading a team of designers covering Alexa Communication, Social, Kitchen, Home Productivity, and Alexa for Everyone, and she's also the editor and author of the book, "Ubiquitous Voice: Essays from the Field." Not bad.
She'll be joining us at VOICE & AI to attempt to answer a simple question that may not have an easy answer. And that question is: What is the role of the screen in a voice-first interaction? And she breaks that down into two sub-questions: Where might it help the interaction? And where might it hurt the interaction? Hence the title of her talk.
Lisa will draw upon her vast experience with conversation design and multimodal interfaces to present informed guidelines to be followed when creating multimodal voice experiences - specifically voice-first experiences augmented with visual elements (the screen).
There's no shortage of devices equipped with both microphones and screens - multimodal experiences appear to be the inevitable way forward. If you want to learn how to create multimodal experiences that grow your brand, you'll want to be in the audience for this one.
Panel: Conversation Design in the Age of Generative AI
Aziz Ghadiali, founder and CEO of VoiceXD, will share the stage with Eros Marcello, Founder and Chief Product Architect at black dream ai, Tom Hewitson, Founder of labworks.io, and Lisa Falkson, Senior Conversation Designer at Amazon, for an enlightening panel discussion on the impact of generative AI on conversation design.
There's little doubt that the rise of generative AI has had a transformative effect on conversation design and its role within the AI toolbox. This multidisciplinary panel will delve into what it takes for conversation designers to build compelling AI assistants that facilitate successful interactions - and how to measure that success.
The panel will also draw upon their collective experience to provide insights into the future conversation design while focusing on the continuing evolution of the conversation designer's role, tasks, and responsibilities.
It's bound to be as poignant as it is practical. If you'd rather be in the driver's seat than wait for a ride, you'll want to be in attendance for this panel.
Conversation design has become a bit of a buzzword lately - and that only highlights its relevance. As AI continues to grow, it is transforming industries across the board - including the AI industry itself. Case in point: conversation design is currently in a bit of a Renaissance - taking in the current AI landscape - with its greater complexity and sophistication - to learn to operate - and design - within the new state of the art.
And it will be transformative.
At Modev, we're proud to be part of this transformation by shining a light on some of the industry's top talent. We hope to see you at VOICE & AI, September 5th through the 7th, at the Washington Hilton in Washington, DC. You can register today at https://www.voiceand.ai/.
Modev believes that markets are made and thus focuses on bringing together the right ingredients to accelerate market growth. Modev has been instrumental in the growth of mobile applications, cloud, and generative AI and is exploring new markets such as climate tech. Founded in 2008 on the simple belief that human connection is vital in the era of digital transformation, Modev makes markets by bringing together high-profile key decision-makers, partners, and influencers. Today, Modev produces market-leading events such as VOICE & AI and the soon-to-be-released Developers.AI series of hands-on training events. Modev staff, better known as "Modevators," include community building and transformation experts worldwide.