Название: Moral Codes: Designing Alternatives to AI
Автор: Alan F. Blackwell
Издательство: The MIT Press
Год: 2024
Страниц: 239
Язык: английский
Формат: pdf (true)
Размер: 10.1 MB
Why the world needs less AI and better programming languages. Decades ago, we believed that robots and computers would take over all the boring jobs and drudgery, leaving humans to a life of leisure. This hasn't happened. Instead, humans are still doing boring jobs, and even worse, AI researchers have built technology that is creative, self-aware, and emotional—doing the tasks humans were supposed to enjoy. How did we get here? In Moral Codes, Alan Blackwell argues that there is a fundamental flaw in the research agenda of AI. What humanity needs, Blackwell argues, is better ways to tell computers what we want them to do, with new and better programming languages: More Open Representations, Access to Learning, and Control Over Digital Expression, in other words, MORAL CODE. Blackwell draws on his deep experiences as a programming language designer—which he has been doing since 1983—to unpack fundamental principles of interaction design and explain their technical relationship to ideas of creativity and fairness. Taking aim at software that constrains our conversations with strict word counts or infantilizes human interaction with likes and emojis, Blackwell shows how to design software that is better—not more efficient or more profitable, but better for society and better for all people. Covering recent research and the latest smart tools, Blackwell offers rich design principles for a better kind of software—and a better kind of world. In an introductory class in programming, students learn to code. They will be taught the syntax and keywords of some programming language notation, perhaps Python or Java, and will learn how to translate an algorithm into the conventional idioms of that particular programming language. Introductions to Machine Learning are not yet quite as familiar or popular as learn-to-code initiatives, but they are equally accessible to children—I have a friend who teaches Machine Learning methods to eight-year-olds in an after-school club.
