Fair AI practices can boost workplace creativity

New research published in Administrative Sciences has found that employees are more likely to use artificial intelligence creatively when they believe workplace AI systems are fair, transparent and ethically managed.

The study, titled "When AI Fairness Shapes Creativity: The Mediating Role of Attitudes Toward AI Across Gender," surveys data from 214 highly skilled employees working in technologically advanced environments to find out how perceived AI fairness influences employee creativity, whether attitudes toward AI help explain that relationship, and whether gender changes the effect.

AI fairness emerges as a driver of creative work

The study defines perceived AI fairness as employees' belief that their organization applies AI-driven practices in a fair, transparent and consistent way. This includes whether AI systems are seen as unbiased, whether their use is clearly communicated and whether employees believe they are treated equitably when AI is used in workplace decisions or tasks.

The study frames creativity as a human process that can be strengthened through collaboration with AI. Employees remain central to the creative process by interpreting, evaluating, selecting and adapting AI-generated suggestions into useful outcomes. Debates over generative AI often split between excitement over faster idea generation and concern that AI may weaken independent thinking or produce more uniform outputs. The study acknowledges that AI can work in both directions. It can support ideation, problem-solving and creative exploration, but it can also encourage overreliance, standardization and reduced originality if workers use it passively or distrust the system.

The findings show that perceived AI fairness has a positive and significant relationship with creativity. Employees who see AI practices as fair are more likely to feel safe experimenting with new approaches. They are also more likely to view AI as a tool that supports their abilities rather than as a mechanism of control, surveillance or replacement.

That relationship is explained through Social Exchange Theory, which suggests that employees respond positively when they believe organizations treat them fairly. In AI-enabled workplaces, fair and transparent technology practices can reduce anxiety and increase trust. Workers may then reciprocate by investing more effort in creative tasks, taking thoughtful risks and engaging more deeply with problem-solving.

The study warns that unfair AI practices can have the opposite effect. If employees believe AI systems are biased, opaque or imposed without explanation, they may resist using them or disengage from creative work. In that sense, fairness is not only an ethical concern. It becomes a practical factor that can influence whether organizations capture the creative benefits they expect from AI adoption.

Positive attitudes toward AI help convert fairness into creativity

The study majorly focuses on attitudes toward AI as a mediating factor. Attitudes toward AI refer to employees' positive or negative evaluations of the technology, including acceptance, enthusiasm, skepticism, anxiety or fear. The research finds that perceived AI fairness positively influences employees' attitudes toward AI. Workers are more likely to accept and use AI when they believe it operates within a fair organizational environment. When systems are transparent and ethically grounded, employees are more likely to see AI as useful and less likely to interpret it as a threat.

Positive attitudes toward AI were found to significantly increase creativity. Employees who view AI favorably are more willing to integrate it into their daily work, use it to test alternatives, generate ideas and improve problem-solving. This supports the Technology Acceptance Model, which holds that people are more likely to adopt a technology when they see it as useful and easy to engage with.

The model shows that attitudes toward AI partially mediate the relationship between perceived AI fairness and creativity. In practical terms, fairness boosts creativity both directly and indirectly. It directly creates a work environment where employees feel supported, trusted and psychologically safe. It also indirectly strengthens creativity by improving how employees think and feel about AI.

The partial mediation finding is important. It means fairness itself remains powerful even when attitudes toward AI are accounted for. A fair organizational climate may stimulate creativity independently by increasing confidence, trust and engagement. At the same time, positive attitudes toward AI help translate that fair climate into active, creative use of the technology.

The study's data support this structure. Perceived AI fairness had a significant positive effect on creativity. It also had a significant positive effect on attitudes toward AI. Attitudes toward AI, in turn, had a significant positive effect on creativity. The mediation effect was present but partial, indicating that fairness and attitudes work together rather than one replacing the other.

The sample helps explain the results. Respondents were highly educated, with nearly 60% holding a master's degree or higher. Many worked in advanced sectors such as education, engineering and business. Most came from North Africa and the Middle East, with smaller representation from Asia, Europe and North America. The study notes that such respondents may be relatively prepared for AI adoption, which could make fairness and attitudes especially relevant in shaping creative engagement.

Gender did not change the fairness-attitude link

The study also tested whether gender moderates the relationship between perceived AI fairness and attitudes toward AI. Earlier research has suggested that men and women may differ in how they evaluate AI systems, with women sometimes showing greater concern over algorithmic bias, accountability and ethical risks, while men may report greater confidence in using AI technologies.

The findings did not support that expectation. Gender did not significantly moderate the link between perceived AI fairness and attitudes toward AI. In other words, men and women in the sample responded similarly when judging AI fairness and forming attitudes toward AI.

The result challenges assumptions that gender differences remain central to AI adoption in skilled workplace environments. The study suggests that broader access to AI tools, greater digital exposure and rising technological competence may be reducing traditional gender gaps. In settings where both men and women have advanced education and regular contact with digital systems, organizational fairness may matter more than gender in shaping attitudes toward AI.

The finding also points to a possible shift toward a more inclusive digital workplace. As AI becomes embedded in professional routines, employees may evaluate it more through practical usefulness, fairness and organizational context than through gender-based expectations. The author argues that this may reflect an emerging convergence in how men and women engage with AI technologies.

Further, the study urges caution, noting that the sample was specific: highly skilled, technologically exposed and largely drawn from the MENA region. The result may not apply equally across all workplaces, sectors or cultural contexts. Less digitally prepared groups, organizations with lower AI literacy or settings where gender inequalities remain stronger may show different patterns.

If organizations want AI to enhance creativity, they must do more than deploy advanced tools. They must build fair, transparent and participatory systems around those tools. Training programs can help employees understand AI functions, reduce misconceptions and build confidence. Co-design initiatives can involve employees in AI integration, strengthening perceptions of fairness and ownership. Open communication channels can allow workers to raise concerns, seek clarification and contribute ideas.

The study also highlights the need for ethical governance. Employees are more likely to use AI creatively when they believe systems are not biased, not hidden and not designed to undermine their autonomy. Clear explanations, fair procedures and responsible implementation can turn AI into a creative partner rather than a source of anxiety.

It is worth mentioning that the research relied on an online questionnaire, which means participants reported perceptions rather than being observed in real AI-use situations. Its cross-sectional design also prevents firm causal conclusions. The sample's cultural and professional profile may limit generalizability. Future studies could use longitudinal designs, real workplace settings and comparisons across cultural clusters to test whether the same relationships hold in broader populations.