As AI systems change our world of work, they will also have an impact on how organizations hire, train, and educate people. How will job profiles and employee requirements change? And what are organizations that are particularly good at mastering these changes like?
Which tasks do we want to transfer to AI systems and which will we prefer to do ourselves in the future? A viral tweet by book author Joanna Maciejewska sums up this dilemma: “I want AI to do my laundry and dishes so that I can do art and writing; not for AI to do my art and writing so that I can do my laundry and dishes.” Of course, what she writes about her life as a creative freelancer applies just as much to the world of work as a whole – and to the healthcare industry in particular.
How do we want to make best use of AI systems, and how should organizations do that? Do we want to use AI to relieve us of the most complex, difficult, but possibly most important tasks? Should AI accompany a person in their final hours or communicate a diagnosis to a sick person? Or would we rather not use AI for routine tasks and tedious activities to free up time and energy for bigger, more important, and more fulfilling tasks? To be able to write a book like Joanna Maciejewska, research in peace, think about a relevant and meaningful innovation in our own industry, or have a little more time for a seriously ill patient.
Every 30 seconds a nurse applies for a job via LinkedIn
The pace at which AI is changing the way we work is increasing fast. And there is no sign of the speed and scale of these changes slowing down any time soon. Aneesh Raman, Vice President and Workforce Expert at the business network LinkedIn, describes the era we are currently in as the era of organizational dynamism. At first glance, it may seem unusual for someone from a career network like LinkedIn to be taking part in a panel discussion on healthcare and medical technology, as Raman did at this year’s SXSW innovation and technology conference. But it actually makes perfect sense because who could be a better seismograph for the current upheavals in the world of work than LinkedIn with around a billion members worldwide?
While social networks such as Facebook or X (formerly Twitter) depict a social graph, i.e. the personal networks of people in the digital world, LinkedIn provides an unprecedented view of the economic graph, an extremely granular view of the labor market. “Every minute, two nurses apply for a job via LinkedIn,” said Raman during a panel discussion entitled Will AI Replace Healthcare Workers? No, But It Will Turn Them Into Tech Workers. “Many of the most booming occupations we see right now are in the healthcare sector,” he continued. “Six of the top ten most highly demanded occupations globally are now in the healthcare sector.”
LinkedIn is seeing a higher frequency of job changes and rapidly changing job profiles across all sectors. As the healthcare industry is heavily people-centered, in comparative terms it is still less susceptible to the upheavals caused by the latest AI wave. “It’s much more dramatic for software developers,” says Raman, whose previous jobs included writing speeches for U.S. President Obama. “There, 96% of jobs are at risk from AI.” But AI is changing the requirements and job profiles in the healthcare sector, too. There, the skills demanded in job adverts have only changed by 1% since 2015. But that will rise to 68% over the next five years, LinkedIn’s research department forecasts.
AI – more of a job transformer than a job killer
If new and changed skills will be required for a job in around two thirds of cases, this will mean completely new roles for some and understandably scare many people. That is understandable. After all, as with many technical upheavals, the negative effects on existing job profiles are evident first. There were good reasons why German weavers and spinners joined forces in the mid-19th century to protest against the introduction of mechanical looms – and often even destroyed them as they saw themselves being deprived of their livelihoods. The new job profiles created by such technological change often only become evident after the losses have occurred, and are often less tangible and dramatic.
In the field of AI, the situation is likely to be again different. Even if the fear of AI as a job killer is repeatedly invoked, many job profiles are likely to change through the use of AI systems rather than being completely replaced by them. Aneesh Raman is certain that “the dystopian fear that AI could destroy jobs in the healthcare sector is unfounded. If anything, it will expand the occupational fields. And there will be completely new job profiles such as head of AI, machine learning engineer or computational biologist.”
But what do these rapidly changing requirements and job descriptions mean for the organizations themselves? How can companies better prepare for and manage the increasing speed and dynamism confronting the labor market, for example? One hypothesis, which was also put forward by SXSW speaker Ian Beacraft, is that some routine tasks as well as some specialized ones will be taken over by AI systems. So, we humans will be able to again concentrate on our human skills, with soft skills again becoming more important, and work humanized.
Understanding change with the management mixing console
Of course, organizations need not only to allow this to happen, but wherever possible, actually encourage it. Yet to this end, it is necessary to understand from an organizational sociology perspective how an organization’s division of labor functions and on which internal relationships individual positions or job profiles are based in an organization.
The Management Mixing Console ©Metaplan
The Metaplan management mixing console is a tool that can be used to illustrate the structures of an organization very well. With regard to the above-mentioned changes brought about by AI, you can look at the console by asking the following questions, for example:
- How will communication channels (where decisions are made) change as a result of the changed task areas and requirements? What will it mean, for example, if sales force staff are instructed by an AI system about what topics they should address in their next customer interaction, and how that should be done? What will change in a hospital if there is an automated AI documentation system and or an AI system that generates ‘suspected of…’ diagnoses based on recognized patterns?
- How will new AI technologies affect an organization’s programs (which decide what is right or wrong)? Will this change the rules on who is allowed to do what and when an error or breach of rules has occurred? For example, for which use cases may AI tools be used and where not – for safety reasons or ethical concerns? In which cases will it be permitted to revise a positive AI decision to validate an antibody target and stop a study program? What accountability / responsibility rules will be needed and how will they be exercised?
- What changes will result at a personnel level? This aspect of organizational dynamism is perhaps the most serious right now. What will happen to recruitment criteria and career paths? How can an organization ensure that existing staff become more digitally capable and digitally savvy? What types of work will AI structures take over and what will happen to staff who have been doing that work up to now? How will the culture or the way of thinking change in an organization change as a result of an increasingly digital-savvy workforce?
- The impact on the management mixing console should naturally not be underestimated, either. What impact will AI systems and other new technologies have on leadership requirements? If we define leadership as successfully exerting influence at critical moments, then AI can set or necessitate leadership impulses. Today’s large language models generate two different answers to the same question. How will these models be used in organizations if what they say can cause surprises?
LinkedIn manager Aneesh Raman chose a different image than the mixing desk to illustrate how AI can ensure that over time, occupations will change completely without the people behind them necessarily having to look for a new job. Raman suggests forgetting about job titles for a moment and instead sorting the individual tasks a job description involves into three buckets: bucket one for tasks that increasingly will be entirely performed by AI; bucket two for tasks you will be able to do yourself with the help of AI – and hopefully better, faster or both; and finally, bucket three for tasks that will only be able to be carried out by humans. This third bucket will be much more important in the future than it has been to date.
Nevertheless, it is not just this shift but also the other changes brought about by organizational dynamism that will demand a new culture. That is why at LinkedIn’s parent company Microsoft, the era of learn-it-alls commended some time ago. According to Microsoft CEO Satya Nadella, the era of know-it-alls is finally over and that of learn-it-alls has dawned. This is probably a very good development because to be honest, nobody ever liked know-it-alls.
