Innovation and knowledge in sustainable supply chains: how ethical climate and training reduce greenwashing risk

Sustainability has become a central part of business strategy over the past few decades, and organizations have begun to recognize the strategic importance of integrating environmental concerns into operations and branding. Businesses are also using innovation and knowledge management to develop green programs that not only improve companies’ environmental performance but also increase competitiveness and corporate image. With stakeholders, such as consumers, regulators, investors and employees, raising their expectations and requiring companies to be more environmentally responsible, companies embrace green policies to enhance their perception and positioning in the market (Sivapalan et al., 2024). At the same time, there has also been a reverse trend: so-called greenwashing or the spread of false or inflated information about environmental efforts (Ren et al., 2024). Greenwashing undermines stakeholder trust, weakens corporate credibility and harms sustainable development goals (Alabdali et al., 2024). Given its complexity, scholars began scrutinizing greenwashing across organizational culture, ethical leadership, innovation, knowledge and supply chain governance (Lv et al., 2023).

The antecedents of greenwashing have been studied extensively in the empirical literature, with particular emphasis on internal organizational factors (Dutta et al., 2023). It is shown that the ethical climate (EC) shapes employees’ organizational norms and behaviors, which in turn affect their tendency to engage in environmental deception and their ability to adopt genuine sustainability programs (Gao and Wei, 2025). Shi et al. (2023) categorized ECs into law-and-code, caring and instrumental types and reported that principled climates were negatively associated with unethical behavior. Similarly, Zhang (2023) found that ethical leadership and adherence to moral standards reduce the likelihood of exaggerating environmental efforts. Empirical evidence has also been derived on sustainability training, which is positively related to better ethical behavior. Free et al. (2024) have shown that employees who obtained sustainability-oriented training were more inclined to implement authentic sustainable practices and resist the superficial training, which is supported by Xu et al. (2023), who have highlighted that the practical training allowed the employees to identify discrepancies between environmental statements and operational reality and oppose them.

Mediator variables also provide a more detailed explanation of the pathways linking organizational antecedents to the risk of greenwashing (Muzaffar et al., 2024). Green skepticism serves as a processing mechanism, enabling employees and customers to remain vigilant in identifying inconsistencies in sustainability messages (X. Hu et al., 2023). In the same vein, a shared vision of environmental goals has also been linked to improved sustainability performance and reduced misreporting (Zhang et al., 2023). The intensity of supplier monitoring has become an important mechanism for imposing accountability in multitier supply chain networks (Lee et al., 2024). All these results indicate that EC, sustainability training and mediating mechanisms are essential to assessing organizational greenwashing behavior.

Although this has been achieved, the literature has some significant gaps (Verdecchia et al., 2023). The analysis of EC and sustainability training has been mainly conducted separately, with little or no study of their interactive or sequential impacts on the risk of greenwashing (Abbas and Khan, 2023). The role of EC in greenwashing in complex organizational environments is underresearched (Nygaard and Silkoset, 2023), and the translation of knowledge gained through training into resistance against deceptive behavior is poorly studied (Jin et al., 2023; Štreimikienė and Ahmed, 2021). Furthermore, little empirical attention has been given to the interaction between internal factors and cognitive and behavioral processes, i.e. skepticism, shared vision and supplier monitoring. Employee skepticism, green shared vision (GSV) and supplier monitoring are among the other mediators that are essential in translating internal values and training into visible decreases in greenwashing (Yan et al., 2023; Kumar et al., 2024; S. Hu et al., 2023; Caglar et al., 2024). A synthesizing framework that considers these mediating mechanisms is therefore required to provide an all-inclusive explanation of greenwashing prevention (Lee et al., 2023).

This paper implements the Theory of Planned Behavior (TPB) (Khan et al., 2023) and the Stakeholder Theory (Chen and Dagestani, 2023) to reveal the way in which internal organizational values and beliefs are converted into behavioral results. According to TPB, individual norms, attitudes and perceived behavioral control influence intentions and behaviors (Vangeli et al., 2023). The EC in organizations influences normative beliefs and shared intentions, thereby reducing the prevalence of unethical practices, including greenwashing (Shang et al., 2024). Sustainability training improves perceptions of control over ethical behavior, thereby promoting the application of the TPB to environmental decision-making. The Stakeholder Theory underscores the importance of aligning the organization’s behavior with the interests of various actors, such as consumers, regulators and employees (Chen et al., 2024). Organizations can address stakeholder expectations for transparency and authenticity by integrating sustainability into training, culture and supply chain monitoring (Gourevitch et al., 2023). Mediating variables – green skepticism, shared vision and supplier monitoring – are the ways that the internal ethical and strategic drivers are operationalized. This study, therefore, explores the direct effects of EC and sustainability training on the risk of greenwashing and the mediating effects of skepticism, shared vision and monitoring rigor in a knowledge-intensive and innovation-based organizational setting.

Stakeholders are increasingly demanding more from organizations regarding their sustainability practices and environmental statements (Santos et al., 2024). In turn, a significant number of companies are interested in cutting an impressive green image and are willing to engage in false or exaggerated communication, so-called greenwashing (Sivapalan et al., 2024). It is not just a problem with communication but indicates the underlying aspects of the organization, culture and governance (Alabdali et al., 2024). Knowledge integration and innovation are essential factors in genuine sustainability practices, as organizations that integrate innovative methods with knowledge-based strategies are better positioned to avoid deceptive practices and establish environmental legitimacy.

One key factor in greenwashing is an enterprise’s moral climate. Empirical studies have shown that ECs have a substantial impact on staff behaviors and administrative decision-making, which, on the one hand, affects the probability of adopting a deceptive culture and, on the other hand, can promote genuine sustainability activities (Gao and Wei, 2025). Caring or principle-based EC settings exhibit fewer tendencies toward greenwashing, as ethical norms are internalized and guide daily decision-making (Zhang, 2023). On the other hand, there is a positive correlation between instrumental ECs, in which decisions are made primarily based on self-interest or short-term benefits and the risk of greenwashing, as employees are unlikely to feel a strong commitment to the organization’s actual sustainability (Xu et al., 2023).

Another important factor is sustainability training. Research on sustainability training as a structured means of increasing employees’ environmental and ethical awareness and their innovative problem-solving (Lee et al., 2024) will prepare them to reject false environmental claims (X. Hu et al., 2023). Sustainability training programs are not only about regulatory compliance but also encourage ethical reasoning, critical thinking and the application of knowledge in sustainability contexts (Abbas and Khan, 2023; Kumar et al., 2024). Empirical research shows that organizations that invest in practical sustainability training are less likely to be exposed to reputational risk, as employees are better positioned to identify, question and report greenwashing practices (Caglar et al., 2024; Free et al., 2024).

Greenwashing has direct and indirect effects on internal and mediating factors. Green skepticism is a cognitive filter that empowers employees to approach sustainability communications critically, preventing them from being deceived by false statements (X. Hu et al., 2023). Likewise, a shared green vision brings workers together around a set of sustainability objectives, enhancing environmental performance and minimizing misreporting (Zhang et al., 2023). Such mediators ensure that EC and training are translated into behavioral consequences, thereby strengthening the organization’s transparency and accountability. Climate-based ethics grounded in integrity and similar values not only discourage greenwashing but also strengthen knowledge sharing, innovation and ethical leadership, thereby benefiting sustainable practices (Lv et al., 2023; Dutta et al., 2023). Sustainability training can supplement these effects by increasing environmental knowledge, critical reasoning and a practical approach, thereby making them less vulnerable to false practices (Shi et al., 2023; Muzaffar et al., 2024). As a result, companies with sound ethical operating cultures and strong sustainability education are more likely to have internal control mechanisms that mitigate the risk of greenwashing, such as shared responsibility, transparency and ethical behavior (Zhang et al., 2023):

A growing empirical literature has examined the role of green claims skepticism as a key psychological mechanism in predicting consumer and worker responses to organizational sustainability communication (Verdecchia et al., 2023). Understanding this skepticism not only advances knowledge but also fosters innovation in the design of transparent and credible sustainability communication strategies. Green claims skepticism is typically defined as a tendency to doubt the truthfulness, motives or credibility of environmental information provided by firms (Nygaard and Silkoset, 2023). Empirical research, including that by Yan et al. (2023), indicated that skeptics would be less likely to believe environmental messages outright and would critically analyze business claims of sustainability even more so. Similarly, research by S. Hu et al. (2023) supports the notion that greater skepticism can act as a buffer against fraud involving fake environmental claims. In internal organizational settings, this cynicism is enacted when employees become more critical of virtue-signaling sustainability campaigns or communications strategies that do not seem to translate into concrete actions for the environment (Lee et al., 2023). Research has also indicated that organizational ethical culture and learning environments also foster skepticism (Chen and Dagestani, 2023). For example, institutions that encourage transparency, accountability and education about sustainability reduce the blind acceptance of corporate communications and enable employees to more easily detect inconsistencies in environmental reports (Shang et al., 2024). Therefore, the existing literature supports the notion that green skepticism is a prevention strategy against greenwashing through ensuring critical examination and enforcing authenticity in environmental communication, while simultaneously advancing innovation and knowledge in sustainable business practices (Anwar et al., 2025).

According to empirical studies, a notable point is that green claim skepticism is not merely a passive trait, but an active filtering mechanism that influences individuals’ responses to sustainability communications and the risk of greenwashing perception, thereby fostering innovation and knowledge in assessing environmental claims (Gourevitch et al., 2023). When employees are skeptical, they are less likely to believe environmental stories are genuine and to endorse superficial or ambiguous practices, thereby reducing the risk of greenwashing within the organization (Sivapalan et al., 2024). Furthermore, skepticism can act as a mediating factor between organizational internal factors and the expression of greenwashing risk (Lv et al., 2023). In addition, skepticism would serve as a mediating variable between organizational intrinsic variables and the expression of greenwashing risk (Gao and Wei, 2025). For instance, in organizations with an excellent EC, workers are likely to adopt skeptical values and moral accountability, which creates a good level of skepticism to disqualify deceptive information (Free et al., 2024). Such acquired skepticism then becomes a potent mediator of the impact of both EC and sustainability training on greenwashing risk, as it equips individuals with knowledge and innovative capacity to act as guardians of truth, advocating only for good, consistent and genuine environmentalism (X. Hu et al., 2023). Empirical evidence, therefore, supports the contribution of skepticism as both a direct and mediating variable in reducing greenwashing behavior:

GSV is the degree to which organizational members agree and understand environmental objectives and sustainability principles (Lee et al., 2024). Empirical evidence highlights its role not only in reinforcing environmental integrity but also in cultivating innovation and knowledge sharing across the organization (Anwar et al., 2025), thereby discouraging unethical environmental actions, such as greenwashing (Kumar et al., 2024). Nygaard and Silkoset (2023) have established that when organizations have a GSV, employees are motivated to engage in activities toward long-term environmental goals, leading to genuine sustainability practices. Empirical observations also demonstrate that shared environmental vision improves organizational unity, internal communication and a sense of responsibility for ecological performance (Lee et al., 2023). Shared mind enables workers to glimpse the larger purpose of sustainability activity, deterring shallow or manipulative tendencies (Vangeli et al., 2023). In addition, green shared-vision organizations are open and willing to disclose their environmental performance, thereby reducing the likelihood that they will be perceived as deceitful (Gourevitch et al., 2023). Some studies have revealed that a GSV is not only an enabler of sustainable innovation but also a psychological construct that induces ethical behavior by connecting individual and organizational environmental values (Ren et al., 2024).

Empirical research shows that a GSV plays a vital role in directly influencing and mediating organizational behavior in response to environmental claims (Shi et al., 2023). When employees collectively internalize and pursue a green vision, they not only adopt ethical decision-making but also foster innovation and knowledge exchange, thereby strengthening sustainable practices and reducing the likelihood of manipulative sustainability reporting (Xu et al., 2023). This common aim prevents short-term, manipulative tactics by creating solidarity and a true passion for environmental ends. Furthermore, a shared vision of green can address the tension between an EC and greenwashing by framing environmental issues as a common concern (Zhang et al., 2023). Even in ethically sound organizations, the absence of a distinct green vision may lead to disjointed efforts or inconsistency in green communications. By creating a shared green vision, staff can more effectively translate ethical principles into sustainable actions (Abbas and Khan, 2023). Likewise, sustainability training that is integrated with and supports a GSV can lead to more integrated, value-based actions among employees (Yan et al., 2023). With training embedded in the company’s vision, deeper learning and internalization of sustainability occur, thereby reducing the chances of greenwashing (Caglar et al., 2024). Empirical support thus underlines GSV as a linchpin that both directly reduces greenwashing risk and strengthens the influence of EC and sustainability training through aligned, collective environmental commitment (Chen and Dagestani, 2023):

Supplier monitoring rigor (SMR) has become a key factor in preventing greenwashing, especially in complex supply chains (Chen et al., 2024). Supplier monitoring can be defined as the level of audit, scrutiny and enforcement of compliance with sustainability norms by organizations to ensure internal ethical norms and sustainability practices are used throughout the supply chain (Alabdali et al., 2024). The empirical research findings indicate that stringent monitoring can help companies identify the deviations, stop supplier-based greenwashing and strengthen knowledge transference and innovation through the supply chain (Dutta et al., 2023; Nygaard and Silkoset, 2023). The tracking also indicates the organization’s commitment to sustainability, thereby enhancing credibility and trust among stakeholders (Lee et al., 2024; Gourevitch et al., 2023).

Notably, supplier monitoring acts as a mediating process. EC and sustainability training can only help establish moral intentions and awareness, but until these values are monitored, they may not be implemented, especially on multitier supply chains (Kumar et al., 2024; Lee et al., 2023). Employees trained in sustainability are aware of the need to monitor and can more effectively ensure compliance and identify greenwashing weaknesses (Vangeli et al., 2023; Santos, Coelho and Marques, 2024). This two-way relationship among training, EC and monitoring forms a reinforcing cycle that reduces the risk of greenwashing and enhances innovation, knowledge diffusion and transparency:

Figure 1 shows the conceptual model of the current study.

The research design used in this study was a quantitative study to investigate how the EC, sustainability training effectiveness (ST), green claims skepticism (SK), GSV and SMR influence the risk of greenwashing (GWRisk) in production-based supply chains and thus added knowledge and innovation to the field of sustainable supply chain management. The production companies in the USA were identified, and data were gathered from workers at companies engaged in sustainability reporting, environmental compliance and supplier evaluation. An electronic, structured, guided questionnaire was administered to provide supply chain professionals, including supply chain managers, sustainability officers, logistics coordinators, procurement staff and environmental compliance analysts. In total, 1201 valid responses were obtained, providing a good sample with sufficient statistical power and thus allowing strong inferences.

Constructs were also measured using established scales from the literature, thereby ensuring validity and reliability. All the variables were measured on a five-point Likert scale (1 = strongly disagree to 5 = strongly agree). Table 1 provides an overview of the variables, subdimensions and source studies.

JASP was used to conduct structural equation modeling (SEM), which included mediators and latent constructs. Before SEM, the descriptive statistics and bivariate correlations were calculated to determine normality, distributions and correlations. Cronbach’s alpha was used to measure reliability and average variance extracted (AVE) and composite reliability (CR) were used to measure convergent and discriminant validity. The model’s fit was evaluated using comparative fit index (CFI), tucker-lewis index (TLI), root mean square error of approximation (RMSEA), standardized root mean square residual (SRMR). Mediation effects were tested using bootstrapping to estimate the stability of indirect relationships and to comprehensively evaluate both direct and indirect processes through which EC and sustainability training affect greenwashing risk (Hussain et al., 2019). Harman’s single-factor test was used to assess common method bias, ensuring the survey design did not overstate variance.

SEM was used to test hypothesized causal relationships among EC, ST, SK, GSV, SMR and GWRisk. The model is specified as:

where GWRisk is the greenwashing risk, EC is the ethical climate, ST is the sustainability training, SK is the skepticism, SMR is the supplier monitoring rigor and β₁–β₄ are the standardized path coefficients, with ε as the residual.

To validate and enhance the robustness of the results, a series of machine learning (ML) techniques – Neural Network Regression (NNR), SVM and Decision Tree Regression – was used. In the model of the neural network, the prediction function is provided as:

where f(·) is a nonlinear function that models the neural network using several hidden layers. For SVM regression, the model takes the form as follows:

with x as the input feature vector, ϕ(x) as the kernel-transformed space, “w” as the weight vector and b as the bias. The decision tree model is represented as follows:

where T(x) represents the prediction made by the tree, computed through recursive binary splitting of the input features, these models were tested and trained using 10-fold cross-validation and compared based on performance metrics such as mean squared error (MSE), root mean squared error (RMSE), mean absolute error (MAE) and R-squared (R²) to assess their forecasting accuracy.

Table 2 shows the reliability and validity measures obtained from the partial least squares structural equation modeling (PLS-SEM) analysis. The findings portray high internal consistency reliability and convergent validity across all constructs. Cronbach’s alpha for each latent construct, EC (0.921), sustainability training effectiveness (0.834) and SMR (0.887), exceeds the 0.70 cut-off recommended by Hair et al. (2017), indicating high internal reliability. The CR measures also range from 0.800 to 0.931, further confirming the constructs’ reliability. The AVE for each construct exceeded 0.50, indicating convergent validity. The indicators for each construct exhibit significant factor loadings (all p-values < 0.001), with T-statistics exceeding the critical value of 1.96, indicating high item reliability. These outcomes collectively confirm the measurement model; the constructs are statistically valid and conceptually distinct, allowing for further structural examination

The structural model is robust and reliable, as indicated by the model fit statistics (see Table 3). The SRMR (0.057) of the model is less than 0.08, which indicates the model fits sufficiently, whereas the Normed Fit Index (NFI = 0.616) and chi-square value (Chi² = 11,357.749) are not too large to fit a complex PLS-SEM model. R² values indicate significant predictors, with 79.8% of the variance in greenwashing risk (GWRisk) explained. SMR had the most significant explained variance (R² = 0.777), followed by skepticism toward green claims (SK, R² = 0.588) and GSV (R² = 0.036). These estimates are closely reflected in the adjusted R² values, which indicate the model’s stability. The heterotrait–monotrait (HTMT) ratio was used to measure discriminant validity. The construct pairs are below the 0.85 cut-off, with the most significant value being skepticism and shared vision (0.833), followed by conceptual proximity with a tolerable difference. A construct-to-construct lower HTMT value, such as between GSV and ST (0.406) or SK and EC (0.390), will show strong discriminant qualities. The analysis of the effect size (F2) showed that sustainability training effectiveness (ST) had moderate to substantial impacts on key mediators, such as supplier monitoring (0.289) and skepticism (0.220). Conversely, the EC showed lower F2 values, suggesting that it exerts more subtle indirect effects on GWRisk. All these results substantiate the validity of the structural sufficiency and predictive applicability of the suggested model.

Table 4 presents the path coefficients and the corresponding significance levels for each hypothesized relationship. Out of the 11 hypotheses that were tested, 9 are supported by statistically significant path coefficients (p < 0.05). H1, the direct impact of EC on greenwashing risk (β = 0.045, p = 0.152 and H8 (β = −0.001, p = 0.866) are insignificant, indicating that EC may not directly predict greenwashing and that its impact could be fully mediated. H2, the effect of sustainability training effectiveness on greenwashing risk, is strongly supported (β = 0.539, T = 11.146, p < 0.001), indicating a positive, direct influence. Likewise, H3 through H7, which investigate the impacts of the ethics climate and training on mediators (skepticism, GSV and monitoring of suppliers), are all statistically significant but have lower coefficients (e.g. H5 β = 0.026, p = 0.002). Notably, H9, H10 and H11, which address the specific role of supplier monitoring stringency, are all statistically significant. H9 has a substantial direct effect on greenwashing risk (β = 0.254, p < 0.001), whereas H10 (β = 0.084) and H11 (β = 0.146) affirm its mediating role in the relationship between EC, training and greenwashing risk. The power and importance of these mediation paths reflect the essential role of monitoring mechanisms in reconciling internal values and achieving sustainable outcomes.

Table 5 compares the predictive performance of three ML algorithms: NNR, support vector regression (SVR) and decision tree regression. The neural network achieves the best overall performance, with the lowest MSE (MSE = 0.121), the lowest RMSE (RMSE = 0.348) and the highest R² value of 0.901, indicating strong predictive ability. The decision tree also performs well (R² = 0.879), but the SVR model is slightly behind (R² = 0.859, MAPE = 88.49%).

Figure 2 displays the predictive accuracy measures of three ML regression models: NNR, SVM Regression and Decision Tree Regression. Each subfigure displays observed test values (x-axis) and predicted test values (y-axis). The closer the data points are to the red diagonal line, the more accurate the model’s predictions. In the NNR graph, the points also tend to cluster closely around the diagonal line, providing a strong predictive fit. The model exhibits minimal dispersion, indicating high accuracy and low residual error. The SVM Regression also offers a good fit between predicted and observed values, but with a slightly wider spread, indicating moderate variability in predictions. The Decision Tree Regression shows greater divergence from the diagonal, particularly in extreme-value intervals, indicating lower predictive accuracy. Still, the overall trend is consistent across all three models, validating their ability to predict greenwashing risk based on the input predictors.

Table 6 offers the feature importance scores of each model. Consistent across all models are sustainability training effectiveness, SMR and EC as the leading predictors of greenwashing risk. In the decision tree, for example, sustainability training is the strongest relative importance (28.3%), followed by supplier monitoring (27.8%) and EC (26.0%). Skepticism and a shared green vision contribute less, but still significantly. The SVR and neural network models also corroborate, with increased dropout loss corresponding to the most impactful features, further reiterating the critical roles of behavioral and ethical factors in driving outcomes.

The SEM results are supported by predictive analysis using NNR, SVM and Decision Tree Regression, which help define the relative values of each predictor. Figure 3 indicates that ST, SMR and EC are consistently the strongest predictors across all models. ST has the most significant relative importance (28.3%) in the decision tree, followed by SMR (27.8%) and EC (26.0%), with SK and GSV also significant. These findings are validated by NNR and SVM models, which are evaluated using mean dropout loss and cross-validation. In contrast, NNR performs better because it has a layered nonlinear structure and can extract features. Figure 3 shows the internal structures and performance of the ML models. This NNR model has an input layer, three hidden layers and an output layer that captures the nonlinear relationships among EC, ST, SK, SMR and GWRisk. The SVM model has good training and validation error curves, and overfitting comes with a low cost of constraint violation at the optimal cost. Together, the findings support the causal and prognostic importance of ethical, behavioral and training-related issues in reducing the risk of greenwashing in production-oriented supply chains.

Table 7 presents the additive contributions of individual test cases across models. Negative contributions from EC, supplier monitoring and training in neural network predictions imply that these reduce the risk of greenwashing in some instances (e.g. Case 1: total predicted = −2.19). However, in some cases (e.g. Case 2), the positive effects of these features raise the expected value. Both the SVR and the decision tree models indicate comparable directional effects but with differences in magnitude. These accounts increase model explainability and support the interpretability of sustainability-related forecasts.

To promote genuine sustainability actions, it is necessary to understand the factors that drive greenwashing behavior. With the growing relevance of ESG-related demands in international business environments, companies need to move beyond empty promises to knowledge-based, innovation-focused sustainability initiatives (Anwar et al., 2025). The present research investigated the relationships among the ethical climate EC, sustainability training effectiveness (ST), green claim skepticism (SK), GSV, SMR and greenwashing risk (GWRisk) (Udoh et al., 2025). The combination of SEM and ML in the analysis allowed for capturing both theory-driven causal pathways and predictive nonlinear interactions, providing a comprehensive picture of the organizational antecedents and mediators that control greenwashing.

The confirmation of the reduction in GWRisk by EC and ST supported earlier findings on values-based workplaces and on practical sustainability training that deters deceptive sustainability practices (Caglar et al., 2024; Jin et al., 2023). In particular, ST also demonstrated a substantial impact on mediators, suggesting that the practical training leads to a direct increase in employees’ capacity to identify and counteract greenwashing. Green claim skepticism mediated the EC–GWRisk and ST–GWRisk pathways and it can only influence behavior by showing that ethical awareness and training become behaviorally effective only when employees critically process environmental information (Lee et al., 2024). Likewise, GSV played an intervener in both relations, indicating that a common environmental goal enhances alignment and integrity in communication, which aligns with Vangeli et al. (2023). These results indicate the significance of cognitive and cultural functions in the conversion of ethical and training to visible decreases in greenwashing.

ML analyses were used to support and extend SEM findings, identifying the relative predictive significance of variables and revealing nonlinear interactions. In Neural Networks, SVM and Decision Trees, ST, SMR and EC have been the most potent predictors of GWRisk. Marked, in many cases, ST is more predictive than EC, suggesting that implementing specific training practices can lead to more immediate behavioral change than the gradual change of the entire organizational culture (Ahmed et al., 2021). SK and GSV also had their roles to play, which shows that employee cognition and alignment with the familiar environment are instrumental in limiting greenwashing. SMR scored high in models, in which monitoring intensity is an important control for translating internal ethical norms and training into confirmed supply chain sustainability practices.

ML models also identified interaction effects that could not be determined from linear SEM paths. For example, low SMR and weak EC increased GWRisk, demonstrating the compounding nature of governance gaps. These findings have practical implications: organizations can use predictive models to deploy early-warning systems, monitor employee skepticism and focus on high-risk departments to intervene. The combination of SEM and ML theoretically confirms and has a diagnostic role. SEM elucidates causal processes and ML nonlinear dependencies and leverage points of operation.

The results show that greenwashing can only be mitigated through a multipronged approach. Companies ought to develop ECs, roll out practical sustainability training, encourage innovation, foster knowledge sharing, build environmental awareness and develop a shared green vision. Supplier monitoring is a central control mechanism to realize internal intentions at the external level, especially in multitier supply chains. The combination of SEM and ML complements each other to provide insights into both the predictive and explanatory aspects of greenwashing risk, which can be discussed to inform both researchers and practitioners. Future research can broaden the scope of these models to include sectors and analyze the moderators of the situation, thereby improving the overall generalizability and operationality of the results.

This research also contributes to the theoretical understanding of organizational greenwashing by integrating the EC, sustainability training effectiveness (ST), green claim skepticism (SK), GSV and SMR within a comprehensive framework. The synthesis of these antecedents and mediators expands knowledge on the available literature on sustainable supply chain management. It offers a sophisticated account of how the combination of internal values, competencies and governance mechanisms can reduce the risk of greenwashing (GWRisk). The results confirm the hypothesis that EC and ST are strong predictors of GWRisk, which validates theoretical claims that organizational values and employee capabilities determine sustainability behavior (Anwar et al., 2025). The study explains the moderating effect of employees’ critical analysis of environmental claims, where SK serves as a mediating cognitive mechanism that helps explain the moderation of reputational exposure. Equally, both GSV and SMR are strategic and operational mediators that develop the literature on the mechanisms through which alignment and governance mediate ethical intent into actual risk reduction (Lee et al., 2024; Vangeli et al., 2023).

As a manager, the results provide a practical input. A culture of ethics in organizations should be nurtured at all levels to promote integrity, innovation and knowledge-based sustainability in decision-making. Quality ST programs will give employees the skills to make socially responsible choices and withstand temptations to make false claims of sustainability (Jin et al., 2023). Encouraging GSV between departments sets goals and values, strengthens ethical practice and prevents greenwashing. Instead of being viewed as a threat, stakeholder skepticism can be seen as constructive feedback that can improve sustainability performance and communication. Strengthening SMR provides consistency in sustainability goals across the supply chain and protects the company against reputational losses stemming from supplier misconduct (Caglar et al., 2024; Zhang et al., 2023).

Although it has some contributions, the study has limitations. First, the cross-sectional design is restrictive for causal inference; longitudinal research can provide a dynamic picture of EC, ST and GWRisk over time. Second, this reliance on self-reported survey data can lead to social desirability bias, suggesting that future studies could adopt a qualitative approach (e.g. interviews or field observation) to enhance data validity. Third, the sample was also limited in its geographical scope, which may limit generalizability to diverse regulatory, cultural or industrial settings. Future studies may use cross-country designs to study contextual moderators and industry-specific dynamics.

This paper describes a statistically proven model of the association between internal and external organizational variables and GWRisk. The two antecedents accredited to influence greenwashing reductions are EC and sustainability training, which are supported by three mediators, SK, GSV and SMR, who convert the antecedent translators into a reduction in greenwashing. The combination of SEM and ML demonstrates causal and predictive relevance, indicates nonlinear interactions and supports the strength of the results. Greenwashing undermines the validity and stakeholder trust in an increasingly accountable business environment. This study offers evidence-based solutions to mitigation: the development of ECs, effective ST, the development of knowledge and innovation through GSV, strict SMR and stakeholder skepticism. These findings benefit both theory and practice by advancing sustainable organizational behavior and offering feasible ways to mitigate reputational risk. Finally, the paper highlights the significance of integrating moral foundations, strategic vision and operational excellence to realize true sustainability, paving the way for future research and organizational restructuring.

Data can be provided upon request.

“It is declared that any animal or human medical testing is neither applicable nor relevant for this study.”

The authors declare that there are no potential conflicts of interest concerning the research, authorship and/or publication of this article.”

The Grammarly software was used for grammar, spelling and punctuation checks. However, the researchers do not use any LLM model for writing this paper.

Waqar Akbar, Conceptualization, Formal analysis, Validation. Rana Salman Anwar, Software, Investigation. Rizwan Raheem Ahmed, Supervision, Writing – original draft. Dalia Streimikiene, Project administration, Visualization. Ahmad Naim Che Pee, Data curation; Writing – review and editing. Justas Streimikis, Methodology, Resources.