What do you think?
Rate this book
Your Wit Is My Command: Building AIs with a Sense of Humor
For fans of computers and comedy alike, an accessible and entertaining look into how we can use artificial intelligence to make smart machines funny.
Most robots and smart devices are not known for their joke-telling abilities. And yet, as computer scientist Tony Veale explains in Your Wit Is My Command , machines are not inherently unfunny; they are just programmed that way. By examining the mechanisms of humor and jokes--how jokes actually works--Veale shows that computers can be built with a sense of humor, capable not only of producing a joke but also of appreciating one. Along the way, he explores the humor-generating capacities of fictional robots ranging from B-9 in Lost in Space to TARS in I nterstellar , maps out possible scenarios for developing witty robots, and investigates such aspects of humor as puns, sarcasm, and offensiveness.
In order for robots to be funny, Veale explains, we need to analyze humor computationally. Using artificial intelligence (AI), Veale shows that joke generation is a knowledge-based process--a sense of humor is blend of wit and wisdom. He notes that existing technologies can detect sarcasm in conversation, and explains how some jokes can be pre-scripted while others are generated algorithmically--all while making the technical aspects of AI accessible for the general reader. Of course, there's no single algorithm or technology that we can plug in to make our virtual assistants or GPS voice navigation funny, but Veale provides a computational roadmap for how we might get there.
Most robots and smart devices are not known for their joke-telling abilities. And yet, as computer scientist Tony Veale explains in Your Wit Is My Command , machines are not inherently unfunny; they are just programmed that way. By examining the mechanisms of humor and jokes--how jokes actually works--Veale shows that computers can be built with a sense of humor, capable not only of producing a joke but also of appreciating one. Along the way, he explores the humor-generating capacities of fictional robots ranging from B-9 in Lost in Space to TARS in I nterstellar , maps out possible scenarios for developing witty robots, and investigates such aspects of humor as puns, sarcasm, and offensiveness.
In order for robots to be funny, Veale explains, we need to analyze humor computationally. Using artificial intelligence (AI), Veale shows that joke generation is a knowledge-based process--a sense of humor is blend of wit and wisdom. He notes that existing technologies can detect sarcasm in conversation, and explains how some jokes can be pre-scripted while others are generated algorithmically--all while making the technical aspects of AI accessible for the general reader. Of course, there's no single algorithm or technology that we can plug in to make our virtual assistants or GPS voice navigation funny, but Veale provides a computational roadmap for how we might get there.
312 pages, Hardcover
Published September 7, 2021
5 people are currently reading
27 people want to read
About the author
Tony Veale
10 books2 followersTony Veale is Associate Professor of Computer Science at University College Dublin.
Ratings & Reviews
Friends & Following
Create a free account to discover what your friends think of this book!
Community Reviews
Displaying 1 of 1 review
September 27, 2021
This book had so much potential as a popular science title, but the way it is written limits its audience. Instead of being the fascinating narrative it could have been, it comes across as a textbook lite with a few popular science moments - so it is unlikely to really appeal to either audience. It was certainly hard going for a topic that should have been such fun.
What Tony Veale tries to do is understand and analyse the nature of jokes (the subtitle says sense of humour (well, humor), but the focus is primarily on jokes, which isn't quite the same thing) by looking at how computer software can be made to produce humorous text. There are some insights here, but the trouble is that it is necessary to wade through far too much description of what was necessary technically, which will only be of interest to computer scientists, and even when Veale is discussing what makes something funny, he relies on very stuffy-sounding theory which doesn't really chime with the general reader.
As an example of a limitation, although the subject is supposed to be a sense of humour, there is very little discussion of how this varies between people. For example, I hate The Office-style embarrassment humour, or humour that is dependent on someone getting physically hurt - but I know that some people love this. It would be interesting to know why. Equally, many of the examples in the book that are supposed to be funny really didn't chime with my sense of humour.
The whole exercise is not devoid of interest. I found the way that twitterbots could use material from Twitter to generate potentially funny tweets interesting, for example. But I think it would have been far better either to write a good popular science assessment of the possibilities (for which, I suspect, Veale would have needed a co-author) or to go full textbook on us. The hybrid approach simply didn't work.
What Tony Veale tries to do is understand and analyse the nature of jokes (the subtitle says sense of humour (well, humor), but the focus is primarily on jokes, which isn't quite the same thing) by looking at how computer software can be made to produce humorous text. There are some insights here, but the trouble is that it is necessary to wade through far too much description of what was necessary technically, which will only be of interest to computer scientists, and even when Veale is discussing what makes something funny, he relies on very stuffy-sounding theory which doesn't really chime with the general reader.
As an example of a limitation, although the subject is supposed to be a sense of humour, there is very little discussion of how this varies between people. For example, I hate The Office-style embarrassment humour, or humour that is dependent on someone getting physically hurt - but I know that some people love this. It would be interesting to know why. Equally, many of the examples in the book that are supposed to be funny really didn't chime with my sense of humour.
The whole exercise is not devoid of interest. I found the way that twitterbots could use material from Twitter to generate potentially funny tweets interesting, for example. But I think it would have been far better either to write a good popular science assessment of the possibilities (for which, I suspect, Veale would have needed a co-author) or to go full textbook on us. The hybrid approach simply didn't work.
Displaying 1 of 1 review