IA on the job: adopting because everyone else is?
Adopting AI can become a signal of credibility before its intended use has even been defined. The management question is specific: how should an investment be assessed when the main argument is that competitors, partners or nearby employers are already making it?
What the literature establishes
Paul J. DiMaggio and Walter W. Powell (1983, American Sociological Review) offer a theoretical framework explaining why organisations within the same field tend to become alike, without that convergence necessarily reflecting greater efficiency. They distinguish coercive, mimetic and normative pressures: external requirements, imitation under uncertainty and professional norms. Applied to AI, this framework encourages scrutiny of adoption motives rather than treating diffusion as proof of value.
When adoption also signals legitimacy
John W. Meyer and Brian Rowan (1977, American Journal of Sociology) develop the argument that formal structures can confer legitimacy by incorporating socially recognised models. In particular, they describe a possible decoupling between those structures and actual activities. An AI rollout announcement, a charter or a training catalogue can therefore be visible without working conditions changing: this is an analytical hypothesis, not a finding about AI from their article.
The premature conclusion
It would be excessive to conclude that imitation is always irrational or that all communication about AI is merely symbolic. A major customer’s requirement can be a genuine constraint, and a partner’s experience can provide a useful starting point. The important distinction is between two rationales: remaining a recognised business partner and solving an identified operational problem. (our executive and employee training programmes)
What these studies cannot decide
These classic articles are theoretical contributions, not experimental evaluations of generative AI. They neither quantify the share of investments driven by imitation nor establish that a particular project is pointless. Common practice nevertheless remains weak when an adoption proposal documents competitors’ initiatives but not the work situation that the investment is supposed to change.
A practical check in Fribourg
Within SHR’s « IA on the job » programme, one exercise would be to assess each proposed use through a short form separating external pressure, the local problem and an observable outcome. In Fribourg, cases could involve an agri-food company, a bilingual industrial team and a university service, with criteria specific to each rather than a shared obligation to adopt. Before budget approval, calculate the proportion of proposals containing a documented baseline, a success criterion and a possible solution without AI. This proportion measures decision readiness, not AI’s value; an expectation of modernity can no longer serve, on its own, as an operational justification. To go further: explore the AI on the Job training in Fribourg, or browse our executive and employee training programmes in Switzerland.
In pictures: AI on the Job in Fribourg
- ai on the job
- Fribourg
- research
- training fribourg