A workplace’s adoption of AI does not ensure greater inclusion. Without intentional planning, the technology can instead exacerbate existing disparities among employees, as demonstrated by a company’s practical experience with the approach.
The issue begins with access. Those who already possess technical confidence, extra time, or supportive managers tend to adopt new tools first. A year later, those who fall behind often do so not due to a lack of ability, but because the system was never designed to accommodate them. One company observed that when AI tools were limited to technology teams, the gap widened further. By the time non-technical staff were allowed to participate, they had already fallen behind in both skills and confidence.
However, AI can also dismantle barriers that existed before its introduction. For an employee whose primary language differs from that used in meetings, it can improve communication clarity. For someone with a disability, it can integrate with assistive technologies. For a finance executive managing monthly reconciliations, it can automate repetitive tasks—provided the tool is customized to their specific work, not just demonstrated in a generic training session.
The challenge lies in the fact that the tool alone does not guarantee inclusion. If employees fear judgment for using it, lack the time to learn, or struggle with access, the obstacle merely shifts rather than vanishes. At Excelsoft, the company addressed this by making AI available across all departments, not just technology, from the outset. Support functions, including HR, finance, and sales, were included from day one, ensuring tools designed for engineers were also functional for other teams.
Three early lessons from deploying AI
The first lesson centered on access. Many organizations inadvertently restrict AI use not through explicit policy, but through unnecessary friction: approval delays, unclear processes, or lost requests. As a result, employees who are already confident and well-connected advance, while others disengage quietly. At Excelsoft, governance and security frameworks were established centrally before any rollout. This was not merely about compliance; it was about ensuring access became a right, not a privilege.
The second lesson involved meaningful training. Technical teams often assume that if they master a tool, others will follow. However, a finance executive requires demonstrations of how AI applies to their specific tasks, such as monthly reconciliations, rather than just technical explanations. Training became effective when it connected directly to real work, showing how the tool could streamline existing processes. The objective was not to create experts, but to help employees leverage AI where it delivered tangible value.
Read Also: How digital tools reshaped lasting professional relationships
The third lesson emphasized measuring actual outcomes, not just participation rates, but identifying who was not engaging and why. An AI ideathon open to all employees generated 75 submissions, but the most significant insight came from analyzing where the ideas originated. Sales and business development teams, not just technology, contributed winning concepts. This indicated that problem-solving was not confined to one department; but without the right tools, those ideas often stalled before implementation.
Early data from the company revealed that while openness to participation had improved, the same employees, those with more time, confidence, and managerial support, continued to engage first. The next phase involved direct conversations with non-participants: Did they require additional time, different examples, or encouragement from their managers? Their responses would inform the next steps in support.
In practice, inclusion is not a single initiative. It is a series of decisions, about who gains access, who receives training, and who is given a fair opportunity to demonstrate their capabilities. Without these choices, even the most advanced AI will perpetuate existing divisions.
The critical question remains whether organizations will treat inclusion as an afterthought or a priority. At Excelsoft, the initial response has been the latter; but the effort is far from complete.
Beyond adoption: assessing real inclusion
The company’s approach to inclusion also required rethinking how success is measured. Early metrics, such as participation rates in AI training or tool usage, were insufficient on their own. They indicated who was engaging; but not why others were not. To identify barriers, Excelsoft conducted anonymous surveys and individual discussions with employees who had not adopted AI tools. The feedback revealed recurring themes: some faced time constraints, others lacked confidence in their ability to use the technology effectively, and a few encountered resistance from managers who viewed AI as a distraction from core responsibilities.
Leave a Reply