In August, the top prize in the digital art category of the Colorado State Fair fine art competition went to a man who used artificial intelligence (AI) to generate his submission, “Space Opera Theater.” . He provided the AI, a program called Midjourney, with a simple “prompt” – a textual description of what he wanted. Systems like Midjourney and the similar DALL-E 2 have led to a new role in the age of AI: “fast engineer”. These people can even sell their text wares on an online marketplace called PromptBase.
Midjourney and DALL-E 2 appeared too late to be included in “Working With AI: Real Stories of Human-Machine Collaboration,” by Thomas Davenport and Steven Miller, professors of information systems at Babson College and Singapore Management University , respectively. But the authors note other unpublished titles: director of automation; Senior Director of Content Systems; architect, ethical practice of AI. As the influence of AI expands, its borders with the world of work become more complex. Next step: IA-prompt’s deputy regional director of quality assurance and security.
The bulk of “Working with AI” includes 29 case studies in which business teams integrate automation into a workflow. Each chapter ends with three or four “lessons we learned”. For each study, one or both authors typically interview not only a worker interacting directly with AI, but also the worker’s supervisor, the manager who decided to adopt the technology, the software developer, and AI customers. company. Although they include statistics on, for example, time savings, the reports are largely qualitative.
Working with AI: True Stories of Human-Machine Collaboration
The MIT Press
We may earn a commission when you purchase products through links on our site.
The book is aimed at managers, consultants and students planning their careers. Like none of the above, I have always enjoyed accessible narratives as a diverse investigation into how current technologies can expand the range of human capabilities. Some of the nominations came to sound alike, but the slight level of bland commercial language, like “stakeholder” and “membership,” was positively tolerable.
Early cases lean towards office workers. A system helps Morgan Stanley’s financial advisors customize investment ideas for their clients. Another helps Arkansas State University fundraisers target potential donors and write emails for them. Others suggest life insurance premiums to MassMutual underwriters or help forecast Kroger sales. Either way, humans have the last word. And in many cases, the systems provide explanations for their results, listing, for example, the variables that most influenced a decision.
Later cases break through the walls of the office. A system predicts which field activities will be most dangerous for Southern California Edison workers and recommends precautions. Another highlights neighborhoods where crime is likely to occur and recommends that police patrol the area. (The latter, a form of predictive policing, has raised concerns about biased algorithms. The vendor says it has countermeasures in place, but the book doesn’t elaborate.)
The benefit in most cases is increased efficiency. AI relieves employees of boring and time-consuming work, freeing them to take care of other tasks, such as strategic thinking or customer interactions. The authors spend less time discussing ways machines might work more accurately than humans, though they do mention Stitch Fix, where algorithms help stylists with clothing recommendations. The company’s director of data science notes that it’s generally best not to override AI, whose choices tend to be superior. While algorithms can be biased, humans can too. Stitch Fix’s styling supervisor said the software takes stylists away from their own preferences and closer to those of clients.
Many readers’ first question might be: will AI take my job? Or: Can I replace my expensive employees with AI? The authors’ short answer is: In the near future, no. Rich countries are actually experiencing a long-term labor shortage. And there are still many things that AI (often) cannot do, such as understanding context, managing dynamic parameters, creating a cohesive story, coordinating people, defining a problem, and knowing when to use AI.
The authors include an oft-quoted comment from radiologist Keith Dreyer: “The only radiologists who will lose their jobs because of AI will be those who refuse to work with AI.” The authors elaborate: “If you’re a human being reading this book – and we suspect you are – that means you need to focus on not worrying about being replaced by a machine for ask yourself if you can add value to a job you like where an intelligent machine is your collaborator. Adding value can mean checking machine work to make sure it was done right, making improvements to machine logic or decisions, interpreting machine results for other humans, or performing tasks that the machine cannot or should not do for some reason.”
Yet the authors are unable to predict macroeconomic outcomes. AI will have countervailing effects on the workforce, increasing some positions, reducing others, and modifying or creating still others. In the case studies, the authors found no downsizing, in part because the business was still expanding. But, they note, the book’s portfolio may not be representative. They only included success stories.
Some growing pains appeared. Many workers resisted the change. Some didn’t like being chained to computers. Others hated the AI performance monitoring. Some even mourned the senseless fragmentation they had given in to automation. And solving human-machine collaboration problems can take years. Firms must adopt “collections of new technologies, new business models, new designs of work, new skills, and new financial arrangements,” the authors write.
MM. Davenport and Miller base their book on actual observations and largely avoid far-flung speculation, aside from their statement that “if the singularity does occur, all bets are off.” I imagine a sequel five years from now would be very different. Perhaps the authors could even name an algorithm as a co-author, as two academics did this year when they used a system called GPT-3 to write about GPT-3. Many bets are already reckless.
Mr Hutson is the author of “The 7 Laws of Magical Thinking: How Irrational Beliefs Keep Us Happy, Healthy and Sane”.
Copyright ©2022 Dow Jones & Company, Inc. All rights reserved. 87990cbe856818d5eddac44c7b1cdeb8