Working with AI today as a guide to the future of work

Geoff Hinton, Turing Award winner and AI pioneer, confidently predicted in 2016 that “within five years, deep learning will outperform radiologists” and advised medical schools to “stop train radiologists now”. To which radiologist Keith Dreyer replied, “The only radiologists who will lose their jobs because of AI will be those who refuse to work with AI.”

Dryer is quoted in a new book by Thomas Davenport and Steven Miller and his more educated prediction could have served as the book’s motto. Working with AI: True Stories of Human-Machine Collaboration is based on 29 case studies of human-machine collaboration in companies that are successfully using AI to augment rather than replace their employees. It is an engaging and well-researched survey of the nature of work today and tomorrow.

Davenport and Miller remind us that Hinton was in good company, albeit far off the mark. Predictions for the impact of AI on employment ranged from 50% of all jobs lost to 5% jobs lost (and a net gain of tens of millions of jobs). “We both strongly believe that augmentation is the main impact of AI,” say Davenport and Miller. With Working with AIthey prove their belief is the reality of many businesses today and provide a guide for the future of work.

Companies serving as case studies on the power of augmentation range from financial services (e.g. Morgan Stanley, Mass Mutual, DBS Bank, Radius Financial Group), software (e.g. Salesforce, Intuit), services online/digital (eg FarmWise, Mandiant, Shopee, Stitch Fix), healthcare (eg Stanford Health Care, Good Doctor Technology) organizations in other sectors of the global economy. For each case study, the authors describe the working context of the system of humans and intelligent machines, and include interviews with workers, managers, customers, and technology vendors. They describe the lessons they learned from each case study and offer their views and conclusions on themes common to all cases.

Above all, teamwork is at the root of the successful introduction of AI in the company. “One of the remarkable aspects of all the case study examples in this book is that they involve very complex collaborations between a number of different groups and actors, both within and outside the ‘outside a particular organization’, write Davenport and Miller. It takes a complex and highly coordinated team and it takes time.

“Everyone is a technician” is another important observation. The case studies highlight the transformation of the traditional isolated, isolated “IT department” into integrated technology and tools across the enterprise. This new technological reality is driving the rise of hybrid roles, merging business and technology skills, expertise and experiences. Today, Davenport and Miller write, “The question is not whether to assume a hybrid business-IT role. Rather, it is a matter of choosing what type of hybrid and to what extent. »

The quality, richness, accessibility and usability of data are important success factors for AI programs and supporting systems for data acquisition, integration and management are important success factors in the deployment of AI. According to Davenport and Miller, these support “platforms” and the people who develop and maintain them need to be given the attention and support they deserve.

Another important support mechanism that Davenport and Miller found in their research is what they call “intelligent case management systems.” It is a type of end-user application environment that helps with workflow management, prioritization, recommendations, and data integration. An important aspect of these systems is that humans can take over AI: “One of the great advantages of humans and intelligent machines working side by side is that humans can confirm that an automated decision is ‘reasonable'”.

As for the current and potential impact of AI on jobs, and in particular its impact on entry-level workers, Davenport and Miller describe both negative and positive situations they found in their research. They conclude that “a reduction in entry-level employment opportunities for knowledge workers remains a significant and imminent threat. At the same time, it is still an inconclusive and evolving situation, with multiple countervailing factors and influencing economic forces.

During their investigation, Davenport and Miller found that remote and freelance work, aided and facilitated by smart machines, has become more common. They see both the pros and cons of this type of work, but stress that workers need to interact, talk to each other and socialize, so they can keep learning and coming up with new ideas.

Finally, “what should be done with the limits of AI and human forces?” Davenport and Miller ask. And answer: let humans override AI decisions; AI experts need to fully understand the working context in which their AI solutions are deployed and explain to everyone involved what AI can and cannot do; and everyone should understand that implementing AI is a change management activity.

Working with AI was released towards the end of a year in which the AI ​​”Foundation Models” made headlines for their “creativity”, “sensitivity” and “understanding” and “reasoning” at the human level. So it was heartening for this human (and skeptical) reader to discover that Davenport and Miller “suspect that even with the relentless trajectory of improving AI capabilities, widespread augmentation is here to stay.”