Table 1 Meta competencies and challenges... | Emerald Publishing

Table 1

Meta competencies and challenges identified by participants

Meta-competency	Challenges	Example data
AI-Mediated Problem-Solving and Decision-Making	Contextualizing AI-generated recommendations within strategic goals Difficulty in interpreting AI-generated insights within broader business or organizational strategies. Decision-makers struggle to bridge the gap between AI outputs and long-term goals Over-reliance on AI for decision-making without critical assessment Users tend to accept AI outputs without questioning underlying biases or misalignments, leading to potential suboptimal or unfair decisions	“I know what the AI suggests, but I don't fully understand why it recommends this approach or how it aligns with our overall strategy.” [int] “AI suggests a strategy, but it's not clear how it was derived, making it hard to trust or integrate into strategic planning.” [re] “I trusted AI's recommendation without question at first, but then I noticed that it systematically prioritized certain customer segments over others.” [re] “AI disproportionately favors a specific demographic or market segment, and users only recognize the bias after reviewing long-term results”. [int]
Human-AI Collaboration and Adaptability	Limited transparency in AI decision-making, requiring manual intervention AI's lack of explainability forces human intervention, leading to inefficiencies and trust issues AI-mediated performance metrics misaligning with human contributions AI-generated performance assessments fail to capture qualitative contributions, creating frustration and misaligned incentives	“Since I couldn't see the reasoning behind AI's decisions, I had to manually review resumes to ensure the right candidates weren't overlooked …” [int] “The AI hiring system filters out resumes, but hiring managers manually review candidates due to concerns about AI missing key qualifications.” [int] “KPI system (AI-based) rewards short-term efficiency, but employees working on long-term strategic goals feel undervalued.” [int] “I was optimizing for long-term success, but AI measured my output in a way that didn't reflect my contributions.” [int]
Ethical and Responsible AI Use	Unclear justification behind AI-generated evaluations AI evaluations often provide numerical scores or classifications without sufficient reasoning, leaving employees unable to contest or understand their results AI's role in decision-making creating ethical dilemmas AI is used in high-stakes decisions (e.g. hiring, credit approvals, law enforcement), raising concerns about fairness, bias, and accountability	“I got an AI-generated evaluation, but I had no idea why I was rated lower than my colleague. There was no explanation—just a number.” [int] “People receive AI-generated performance ratings but lack visibility into the factors influencing their scores.” [re] “AI takes over certain decisions, but we don't always know if it's considering all the right factors. There's no human intuition involved.” [re] “AI rejects loan applications but doesn't consider exceptional cases that a human would recognize, leading to unfair outcomes.” [int]
Data Literacy and AI Model Awareness	Difficulty in assessing the quality and biases of AI-generated insights Users struggle to evaluate whether AI models rely on high-quality, representative data or reinforce existing biases AI reinforcing pre-existing biases AI models trained on biased data perpetuate discrimination or social inequalities, requiring careful oversight	“We trust AI's results too much, but how do we know if the data [it learned from] is even relevant?” [int] “AI recommends actions based on outdated or biased datasets, but users lack the expertise to validate its conclusions.” [int] “We use AI-generated reports, but I've noticed sometimes they reinforce existing biases rather than challenge them.” [re] “AI hiring recommendations favor certain demographics due to biased historical hiring patterns.” [int]
Interdisciplinary and Continuous Learning	Gaps between technical and human-centered AI understanding AI experts and domain specialists struggle to communicate effectively, leading to misalignment between AI capabilities and business needs The need for ongoing AI upskilling due to rapid technological evolution AI technologies evolve quickly, requiring continuous learning for professionals to remain competent Challenges in cross-disciplinary collaboration due to differing expertise levels Effective AI implementation requires collaboration between data scientists, business leaders, and domain experts, but communication barriers often arise	“The AI folks get the tech part fine, but they don't always think about how it actually hits people. We need both sides talkin’.” [re] AI engineers optimize an algorithm for efficiency but overlook real-world ethical or practical concerns raised by users. [int] “AI is evolving so fast. What I learned last year is already outdated. If I don't keep learning, I'll fall behind.” [int] “An employee trained on an older AI tool struggles to adapt to a new version with different capabilities.” [int] “We work with data scientists, but sometimes they assume we understand all the technical terms. There's a gap we need to bridge.” [re] “A marketing team struggles to interpret AI-generated insights because the data science team assumes technical knowledge.” [int]

Meta-competency	Challenges	Example data
AI-Mediated Problem-Solving and Decision-Making	Contextualizing AI-generated recommendations within strategic goals Difficulty in interpreting AI-generated insights within broader business or organizational strategies. Decision-makers struggle to bridge the gap between AI outputs and long-term goals Over-reliance on AI for decision-making without critical assessment Users tend to accept AI outputs without questioning underlying biases or misalignments, leading to potential suboptimal or unfair decisions	“I know what the AI suggests, but I don't fully understand why it recommends this approach or how it aligns with our overall strategy.” [int] “AI suggests a strategy, but it's not clear how it was derived, making it hard to trust or integrate into strategic planning.” [re] “I trusted AI's recommendation without question at first, but then I noticed that it systematically prioritized certain customer segments over others.” [re] “AI disproportionately favors a specific demographic or market segment, and users only recognize the bias after reviewing long-term results”. [int]
Human-AI Collaboration and Adaptability	Limited transparency in AI decision-making, requiring manual intervention AI's lack of explainability forces human intervention, leading to inefficiencies and trust issues AI-mediated performance metrics misaligning with human contributions AI-generated performance assessments fail to capture qualitative contributions, creating frustration and misaligned incentives	“Since I couldn't see the reasoning behind AI's decisions, I had to manually review resumes to ensure the right candidates weren't overlooked …” [int] “The AI hiring system filters out resumes, but hiring managers manually review candidates due to concerns about AI missing key qualifications.” [int] “KPI system (AI-based) rewards short-term efficiency, but employees working on long-term strategic goals feel undervalued.” [int] “I was optimizing for long-term success, but AI measured my output in a way that didn't reflect my contributions.” [int]
Ethical and Responsible AI Use	Unclear justification behind AI-generated evaluations AI evaluations often provide numerical scores or classifications without sufficient reasoning, leaving employees unable to contest or understand their results AI's role in decision-making creating ethical dilemmas AI is used in high-stakes decisions (e.g. hiring, credit approvals, law enforcement), raising concerns about fairness, bias, and accountability	“I got an AI-generated evaluation, but I had no idea why I was rated lower than my colleague. There was no explanation—just a number.” [int] “People receive AI-generated performance ratings but lack visibility into the factors influencing their scores.” [re] “AI takes over certain decisions, but we don't always know if it's considering all the right factors. There's no human intuition involved.” [re] “AI rejects loan applications but doesn't consider exceptional cases that a human would recognize, leading to unfair outcomes.” [int]
Data Literacy and AI Model Awareness	Difficulty in assessing the quality and biases of AI-generated insights Users struggle to evaluate whether AI models rely on high-quality, representative data or reinforce existing biases AI reinforcing pre-existing biases AI models trained on biased data perpetuate discrimination or social inequalities, requiring careful oversight	“We trust AI's results too much, but how do we know if the data [it learned from] is even relevant?” [int] “AI recommends actions based on outdated or biased datasets, but users lack the expertise to validate its conclusions.” [int] “We use AI-generated reports, but I've noticed sometimes they reinforce existing biases rather than challenge them.” [re] “AI hiring recommendations favor certain demographics due to biased historical hiring patterns.” [int]
Interdisciplinary and Continuous Learning	Gaps between technical and human-centered AI understanding AI experts and domain specialists struggle to communicate effectively, leading to misalignment between AI capabilities and business needs The need for ongoing AI upskilling due to rapid technological evolution AI technologies evolve quickly, requiring continuous learning for professionals to remain competent Challenges in cross-disciplinary collaboration due to differing expertise levels Effective AI implementation requires collaboration between data scientists, business leaders, and domain experts, but communication barriers often arise	“The AI folks get the tech part fine, but they don't always think about how it actually hits people. We need both sides talkin’.” [re] AI engineers optimize an algorithm for efficiency but overlook real-world ethical or practical concerns raised by users. [int] “AI is evolving so fast. What I learned last year is already outdated. If I don't keep learning, I'll fall behind.” [int] “An employee trained on an older AI tool struggles to adapt to a new version with different capabilities.” [int] “We work with data scientists, but sometimes they assume we understand all the technical terms. There's a gap we need to bridge.” [re] “A marketing team struggles to interpret AI-generated insights because the data science team assumes technical knowledge.” [int]

Source(s): Authors’ own work