AI adoption: should teams change their work pace?
Introducing AI may lead a manager to raise output targets before examining the work that remains. The managerial question is specific: under what conditions can faster execution justify a higher work pace without increasing job strain?
What the literature establishes
Robert Karasek (1979, Administrative Science Quarterly) links psychological strain to the combination of high job demands and low decision latitude. His analysis suggests that workload should not be considered separately from people's ability to organise how they carry it out. For AI adoption, this means distinguishing the production speed offered by the tool from the control people retain over their own pace.
The work automation leaves behind
Lisanne Bainbridge (1983, Automatica) analyses the “ironies” of automation: automating operations can leave humans with monitoring responsibilities and difficult interventions. This classic paper does not measure the effects of generative AI, but describes a mechanism relevant to examining its introduction. A team that produces documents faster may need to devote more attention to the ambiguities and exceptions that remain.
The hasty conclusion
One might conclude that targets should never be raised after introducing AI. These works no more justify that rule than they justify automatically increasing the work pace. Instead, they suggest checking whether additional demands come with greater discretion and whether the remaining work is genuinely less demanding. (our executive and employee training programmes)
What the evidence cannot determine
The observational data used by Karasek do not demonstrate that a particular change in work pace will cause a specific effect in a team using AI. Bainbridge's analysis concerns automation and provides no transferable threshold for acceptable output. Timing text generation is therefore insufficient: checking, rework and the ability to defer a request also need to be observed.
A practical check in Fribourg
As part of the SHR programme “Leading AI adoption in your team”, a Fribourg team in agri-food or bilingual industry could test AI-assisted drafting of non-sensitive internal documents, potentially drawing on methodological support from the local university hub. Over a period defined before the test, compare cases of similar complexity and language requirements, with and without assistance, recording total time to approval, rework time and the actual ability to reprioritise requests. The central measure would be the number of approved cases per total working hour, accompanied by a record of working-time overruns and an assessment of perceived strain. A higher work pace would only be considered if the gain persisted at comparable quality, with no observed increase in overruns or perceived strain. To go further: explore the Leading AI Adoption in Your Team training in Fribourg in the canton of Fribourg, or browse our executive and employee training programmes in Switzerland.
In pictures: Leading AI Adoption in Your Team in Fribourg
- leading ai adoption in your team
- Fribourg
- research
- training fribourg