How HR effectively manages efficiency

Bottom-up instead of strategy: Artificial intelligence is spreading organically in German medium-sized companies. The result: Efficiency gains seep into individual relief instead of systematically flowing into strategic growth. Based on case studies, Caroline Sophie Peitsmeier and Professor Heike Schinnenburg from Osnabrück University show how HR and managers pave the way for individual experiments to become measurable corporate success.

AI challenge: replacement versus augmentation

The availability of freely accessible, powerful AI applications has significantly accelerated technological change in the world of work. AI is considered a key technology that is able to accelerate work processes, support decisions and change entire fields of activity. While earlier automation primarily affected production processes, today there is a different focus. AI seems predestined to take over routines, especially in commercial activities and especially home office-capable tasks: writing texts, answering queries, evaluating data.

The social debate about AI and its effects on the world of work is often heated. Does AI replace humans or does it simply support them? A look at medium-sized businesses shows that the support logic currently dominates. AI is used to make work faster, easier and more pleasant. The major disruption has so far not occurred.

Application in medium-sized businesses

Case studies in three medium-sized companies show that AI has long since become part of everyday work. It is used here primarily in customer service, purchasing and marketing, but also in the HR area. What is striking is less where AI is used, but rather how. A central feature is the acquisition of AI applications through learning-by-doing. The simple operation of the systems makes it easier to get started.

The introduction was almost always bottom-up: employees tried out tools – often privately at first – and adopted working applications into their jobs. Official strategies, clear guidelines or coordinated tool decisions usually only followed later. This makes the application of AI tools fundamentally different from previous software introductions in companies. Instead of large IT projects with training concepts, AI is spreading quietly, quickly and decentrally.

Efficiency gains remain with employees

The use of AI is already paying off for employees. They report time savings, less routine work and better results. AI helps formulate, structure information or prepare conversations. What is missing so far is the organizational perspective. The efficiency gains remain at the individual level. They are neither systematically recorded nor used strategically. Staff cuts are not (yet) an issue, quite the opposite: AI is seen as an answer to the shortage of skilled workers.

Misjudgment of employees

What is remarkable is the low threat perception among employees. Case study company respondents believe AI will support them, not replace them. This is in clear contrast to scientific forecasts, which see a high potential for automation, especially for administrative activities. However, there is a generational difference in employees’ future expectations: While younger employees could hardly imagine that AI could lead to job cuts, older respondents see the potential for automation and also expect savings.

This difference can be interpreted as an assessment based on experience: Older employees are familiar with phases of a difficult labor market with greater competitive pressure, while younger employees have so far benefited from the applicant market in recent years and therefore perceive less of a threat from AI. The calmness is understandable, but also not unproblematic. Because anyone who assumes that fundamentally nothing will change is not preparing for change. This is exactly where a dangerous gap arises between current perception and possible future.

Strategic tasks for HRM

The current situation is both challenging and full of opportunities for HR and personnel development. The AI ​​transformation is currently largely uncontrolled. Employees drive development, HR reacts. The central task is to actively take up this process:

• Create transparency: Where is AI used, for which tasks and with what effects?
• Make efficiency gains visible and consciously decide how they are used.
• Critically question processes: Where can which AI tools be used sensibly?
• Rethink competency requirements: Experienced employees particularly benefit from AI as a sparring partner. Newcomers and career changers need exchange within the team in order to be able to classify and question AI results. Critical thinking, judgment and self-organization are central to the future.
• Reorganize learning: Fewer classic seminars, more exchange formats, best practices and work-related learning.
• Honest communication and encouragement: For many administrative employees, this will involve more than minor changes to their own job. HR will have to accompany an adjustment process and encourage and, if necessary, qualify employees to take on new roles.

AI learning takes place primarily in everyday life, not in the seminar room. This shifts the role of personnel development from being a pure provider of training to a companion of more self-directed learning and change processes. Managers become key players.

In the long term, HR will also have to ask itself uncomfortable questions: How will efficiency gains be used? Which roles are changing? What does AI mean for HR itself when analytics and learning systems are increasingly automated?

Conclusion

The case studies show a clear picture: AI has arrived in medium-sized businesses, is pragmatic, helpful and largely conflict-free. But that is exactly what can be deceptive. The profound structural changes are probably still to come. Companies that just watch now are wasting potential. However, if you consciously design the use of AI, you can gain efficiency, relieve employees and strategically reposition HR. The question is not whether AI will change working life, but who will shape this change.

Also read the following posts: