Transforming Conversational AI: Exploring the Power of Large Language Models in Interactive Conversational Agents

Acquire the knowledge needed to work effectively in conversational artificial intelligence (AI) and understand the opportunities and threats it can potentially bring. This book will help you navigate from the traditional world of dialogue systems that revolve around hard coded scripts, to the world of large language models, prompt engineering, conversational AI platforms, multi-modality, and ultimately autonomous agents.

In this new world, decisions are made by a system that may forever remain a ‘black box’ for most of us. This book aims to eliminate unnecessary noise and describe the fundamental components of conversational AI. Past experiences will prove invaluable in constructing seamless hybrid systems. This book will provide the most recommended solutions, recognizing that it is not always necessary to blindly pursue new tools.

Written in unprecedented and turbulent times for conversational interfaces you’ll see that despite previous waves of advancement in conversational technology, now conversational interfaces are gaining unparalleled popularity. Specifically, the release of ChatGPT in November 2022 by Open AI revolutionized the conversational paradigm and showed how easy and intuitive communication with a computer can be.

Old professions are being disrupted, new professions are emerging, and even the most conservative corporations are changing their strategy and experimenting with large language models, allocating an unprecedented amount of budget to these projects. No one knows for sure the exact future of conversational AI, but everyone agrees that it’s here to stay.

You will:

• See how large language models are constructed and used in conversational systems
• Review the risks and challenges of new technologies in conversational AI
• Examine techniques for prompt engineering
• Enable practitioners to keep abreast of recent developments in conversational AI

Michael McTear is an emeritus professor of Ulster University who has worked in spoken dialogue technologies and conversational AI for more than 20 years. He is author of several books, including Spoken Dialogue Technology (SpringerLink, 2004), The Conversational Interface (SpringerLink, 2016), and Conversational AI (SpringerLink, 2020). Currently Michael is involved in several research and development projects investigating the use of conversational agents in socially relevant projects such as mental health monitoring and home monitoring of older adults. Michael’s main motivation for writing this book is to bring new developments in conversational AI to the attention of conversation designers and other professionals in a clear and accessible manner.

Marina Ashurkina is Product Manager in Generative Assistants Inc., a US-based startup striving to streamline the creation of generative AI assistants. Marina studied linguistics and translation studies, she has over 8 years of experience working with dialog systems, including working in the company api.ai before it was acquired by Google and became Dialogflow, also she had her own consultancy Cherry.ai helping companies build smart assistants and worked on building a multi language voice assistant platform for Huawei. In 2020 Marina created and lectured on a conversation design course to 60 students with a focus on building skills for smart speakers. Besides that Marina is a certified Project Manager Professional, Scrum Master and Product Owner which helps her to set up and drive complex Conversational AI projects.