Designing Experiences for Chatbots
summary
Chatbot projects aim for full natural language understanding but claim that perfect conversational AI is unrealistic today, so teams build hybrid flows combining free text with interactive blocks; practical tools shown include Chatfuel, Amazon Lex, IBM Watson and KeyReply and a platform example runs Hi Bot >1M users/month. Specific advice: be honest about bot identity, design to solve a concrete task (pizza-order tracking worked; fake-human shop bots fail), hire psychologists for language nuance, and prioritize measurable task completion over clever conversation.
AI-written summary · about
chapters (12)
- 0:00Introduction
- 1:45Present
- 6:02Past
- 9:58Main Topic
- 12:40Complex Language Communication
- 13:35Psychology
- 15:22Complex sentences
- 22:00Applications
- 25:31App vs Website
- 29:50Careers
- 31:20Brand
- 32:14Personal assistants
related talks
29mAvoiding the Chatbot Trap Designing AI Interfaces That Unlock Real Business Value
54mIntroduction to Chatbots
50mThe Art of Coding a Conversation: Designing Bots
24mChatbots — Our Conversational Partners Today and Tomorrow
38m3 things about Chatbot (you don't have to learn it the hard way)
43mFrom GUIs to VUIs - How to Design and Create a Chatbot
36mChatbots Past, Present and Future
55mEnterprise Chatbots (using messaging and AI to build novel user interfaces for work)