The Luddites of AI
Friday, January 5, 2024 at 02:30PM
In his book The Perfectionists: How Precision Engineers Created the Modern World*, Simon Winchester reminds us that technological advancement has not always been well-received. The employment of mass production of woolen and cotton textiles in the early 19th century displaced many manual workers who were understandably upset by the loss of their livelihoods. An uprising was supposedly led by a “King Ludd” in opposition to automation, and from that uprising we retain the term “luddite".” Which, of course, means a person who resists technological change. The journalism industry has lately been expressing concern over the impact of artificial intelligence on their business. Social media platforms are being accused of pilfering content from traditional news sources without recompense. (”The Times Sues OpenAI and Microsoft Over A.I. Use of Copyrighted Work”. New York Times, Dec 27. It’s reported that both the number of traditional publications, especially small local newspapers, and the reporters and editors who work for them has substantially decreased in the last few years. And technology is seen as the cause. “... having people, not machines, tell the human stories of the city and the world for other humans to read should stay that way.” New York Daily News summarizes in a recent editorial. Is this truly a threat to good journalism and its effective distribution - or are we experiencing just another case of luddites fearful of being terminated by The Terminator? A major difference between today’s replacement scenario and the days of factory automation is that the work that may be done by AI instead of humans is that it will be intellectual labor, not physical toil. A robotic arm attaching a fender to an auto body does not require insight, empathy, objectivity, or contextual knowledge. A good reporter does. Can AI know and compose basic news stories without boots on the ground, being where the news happens? Or might AI actually be better? Could a bot attend a school board or city council meeting, record it, and then distill the content? Could a bot analyze and summarize congressional actions, supreme court decisions, or presidential activities? Could a bot be embedded in a police force to report on crime or in a courtroom to describe trials? All in a style that is understandable and lacking bias. And could AI provide editorial comment that is actually more balanced than the current humans writing these thought pieces now? Were I programming an editorial AI I’d ask it to always write two editorials on a subject, each reflecting separate sides of an issue, giving readers a more balanced view of issues. I find myself somewhat torn about AI, finding it both potentially promising and perilous. But I don’t want to be classified as a Luddite. For now, AI gets the benefit of the doubt from me. Despite my spell checker just wanting to replace “doubt” with ‘donut” in the last sentence * My grandson recommended this book to me. His senior high school English class teacher assigns books based on each student’s career interests. And Miles is interested in pursuing a career in engineering. The book is much more interesting and readable than the title suggests.
Reader Comments (2)
I think that carefully considering our choices around technology, remaining skeptical in the scientific sense of the word, and asking that technology attend to basic ethics, responsible data management, and transparency does not make someone a luddite. AI is not "inevitable" - it is only as inevitable as our choices allow. If you want to get a taste of what AI journalism is like, just check out AI travel writing. Forget the subtleties of bias, publishers have published travel guides that included places that don't exist! https://jackmontgomery.co.uk/2023/07/10/ai-and-travel-writing/
Hi Geoff,
Your excellent point is well-taken. I agree completely that the expanded use of AI needs to be done carefully and responsibly. I appreciate the reminder.
Doug