Almana : Jurnal Manajemen dan Bisnis Volume 10 Number 1/ April 2026: 136-146 p-ISSN 2579-4892/ e-ISSN 2655-8327 DOI: 10.
36555/almana.
Structural Model of AI-Based Feedback's Impact on Employee Commitment and Creativity to Enhance Organizational Innovation Mari Maryati1.
Tatang Supriyadi1 Universitas Komputer Indonesia1 Indonesia1 *Corresponding email: tatang.
supriyadi@email.
Abstract: The increasing use of artificial intelligence (AI) Article History:
in workplace feedback systems has reshaped how Submitted: March 26, 2026 employees receive performance evaluations and Revised: April 13, 2026 While AI promises efficiency and objectivity, its Accepted: April 23, 2026 influence on employeesAo psychological engagement.
Published: April 30, 2026 creative behavior, and organizational innovation is not yet fully understood.
This study aims to examine how AIbased feedback affects employeesAo work commitment Keywords:
and creativity and how these factors, in turn, contribute to Artificial IntelligenceAeBased organizational innovation.
A quantitative explanatory Feedback approach was employed, using survey data collected Employee Creativity from employees who had experienced AI-supported Organizational Innovation feedback in their organizations.
The data were analyzed Structural Equation Modeling using Partial Least Squares Structural Equation Modeling Work Commitment (PLS-SEM).
The findings reveal that AI-based feedback positively influences both work commitment and employee creativity.
In addition, work commitment and creativity each play a significant role in promoting organizational innovation and jointly mediate the relationship between AI-based feedback and innovation These results suggest that AI-based feedback does not automatically lead to innovation.
instead, its benefits emerge when the technology strengthens employeesAo emotional attachment to their work and supports creative exploration.
In conclusion, this study highlights the importance of designing AI-based feedback systems that are transparent, supportive, and development-oriented in order to fully realize their potential in fostering sustainable organizational Maryati.
, & Supriyadi.
Structural Model of AI-Based Feedback's Impact on Employee Commitment and Creativity to Enhance Organizational Innovation.
Almana : Jurnal Manajemen dan Bisnis.
, 136-146.
https://doi.
org/10.
36555/almana.
INTRODUCTION
The growing use of artificial intelligence (AI) in organizational contexts has reshaped how work performance is evaluated, monitored, and improved.
In recent years.
AI-based systems have increasingly been adopted to provide performance feedback, generate data-driven evaluations, and offer personalized recommendations for employee development.
Organizations expect these systems to enhance efficiency, objectivity, and learning while supporting innovation in an increasingly dynamic business environment (Davenport & Ronanki, 2018.
Jyhnk et al.
, 2.
However, alongside these expectations, there is increasing concern regarding how AI-based feedback influences employeesAo psychological engagement and creative behavior.
This work is licensed under a Creative Commons Attribution-NonCommercialNoDerivatives 4.
0 International License.
https://creativecommons.
org/licenses/by-nc-nd/4.
Almana : Jurnal Manajemen dan Bisnis p-ISSN: 2579-4892 e-ISSN: 2655-8327 Existing research on AI in the workplace presents mixed and sometimes contradictory findings.
On the positive side.
AI has been associated with improved learning processes, higher self-efficacy, and stronger innovation-related behaviors when it supports employees in understanding and improving their work (Zhang et al.
, 2.
contrast, studies on algorithmic management suggest that AI-driven systems may reduce autonomy, weaken emotional attachment, and suppress creativity when employees perceive them as controlling, opaque, or unfair (Li et al.
, 2.
These divergent findings suggest that AI does not inherently promote or hinder innovation.
rather, its effects depend on how employees experience and interpret AI-enabled practices.
From a behavioral and organizational perspective, innovation is not generated by technology alone.
Instead, innovation emerges through employees who are willing to remain committed, invest discretionary effort, and engage in creative problem-solving.
Two mechanisms are particularly central in this process: work commitment and employee creativity.
Work commitment reflects employeesAo emotional attachment and willingness to contribute beyond formal job requirements, while creativity represents the ability to generate novel and useful ideas that can later be implemented as innovations (Anderson et al.
, 2.
Despite their importance, these mechanisms are often examined separately in the literature, leaving the relationship between AI-based feedback, employee attitudes, and innovation outcomes theoretically fragmented.
This study is grounded in Social Exchange Theory (SET), which posits that employees reciprocate favorable organizational treatment with positive attitudes and behaviors (Cropanzano et al.
, 2.
Feedback plays a central role in this exchange relationship because it signals how much the organization values employeesAo contributions and development.
When AI-based feedback is perceived as supportive, fair, and development-oriented, employees are more likely to respond with stronger work Conversely, feedback perceived as impersonal or punitive may weaken the exchange relationship and reduce psychological engagement.
To explain how feedback influences creativity, this study also draws on Feedback Intervention Theory (FIT), which argues that feedback enhances performance and creativity when it directs attention toward task learning and improvement rather than ego defense (Kluger & DeNisi, 1.
AI-based feedback, particularly when timely and actionable, can facilitate iterative learning by helping employees identify gaps, test alternatives, and refine ideas.
This mechanism aligns with recent findings that AI can enhance creativity when designed as a developmental and evaluative tool rather than as a monitoring device (Liu et al.
, 2022.
Zang et al.
, 2.
In addition, insights from technology affordance theory emphasize that the effects of AI depend on how employees perceive and enact the opportunities offered by the technology (Faraj & Azad, 2.
AI-based feedback can afford reflection, learning, and creative exploration, but these affordances are realized only when employees experience the system as enabling rather than restrictive.
This perspective helps explain why similar AI technologies may produce different outcomes across organizational Although prior studies have examined AI adoption in human resource management, most research has focused on isolated relationships, such as AI and creativity or AI and innovation behavior, without integrating work commitment and creativity simultaneously as explanatory mechanisms (Mo et al.
, 2025.
Zhang et al.
Moreover, many studies conceptualize AI broadly, failing to distinguish between AI used for developmental feedback and AI used for algorithmic control, despite their fundamentally different psychological implications (Li et al.
, 2.
This lack of conceptual precision contributes to inconsistent empirical findings and limits theoretical Despite the growing body of literature on artificial intelligence (AI) in organizational settings, several important gaps remain.
First, prior studies have largely examined the Structural Model of AI-Based Feedback's Impact on Employee Commitment and Creativity to Enhance Organizational Innovation Mari Maryati1.
Tatang Supriyadi*1 Almana : Jurnal Manajemen dan Bisnis Volume 10 No.
1/ April 2026: 136-146 impact of AI on employee outcomes in a fragmented manner, focusing on isolated relationships such as AI and creativity or AI and organizational performance, without simultaneously integrating key psychological mechanisms such as work commitment and employee creativity within a unified framework.
Second, existing research often treats AI as a broad and homogeneous construct, failing to distinguish between AI used for developmental feedback and AI applied for algorithmic control, which may produce fundamentally different employee responses.
Third, empirical findings remain inconsistent, with some studies highlighting positive effects of AI on innovation, while others report negative consequences such as reduced autonomy and suppressed These inconsistencies suggest a lack of comprehensive models that explain how and under what conditions AI contributes to innovation.
To address these gaps, this study develops a structural model that positions work commitment and employee creativity as parallel mediating mechanisms linking AI-based feedback to organizational innovation.
By integrating Social Exchange Theory and Feedback Intervention Theory, this research provides a more holistic and humancentered explanation of AIAos role in shaping employee behavior and innovation Therefore, this study contributes to the literature by clarifying the indirect pathways through which AI-based feedback influences innovation, offering both theoretical integration and empirical evidence, particularly within the context of MSMEs in emerging economies.
The purpose of this research is therefore to examine .
the effect of AI-based feedback on work commitment and employee creativity, .
the influence of commitment and creativity on organizational innovation, and .
the mediating roles of commitment and creativity in translating AI-based feedback into innovation.
Through this approach, the study contributes to the business and management literature by clarifying the conditions under which AI-based feedback can meaningfully support sustainable organizational innovation LITERATURE REVIEW The rapid integration of artificial intelligence (AI) into organizational practices has significantly changed how employees receive and respond to performance feedback.
Unlike traditional feedback systems.
AI-based feedback offers real-time, data-driven, and personalized insights that can support employee development and organizational From a theoretical standpoint, this phenomenon can be explained through Social Exchange Theory (SET) and Feedback Intervention Theory (FIT).
SET,
introduced by Peter Blau and further developed by Russell Cropanzano, suggests that employees tend to reciprocate positive organizational treatment with stronger commitment and engagement.
At the same time.
FIT, proposed by Avraham Kluger and Angelo DeNisi, explains that feedback is most effective when it helps individuals focus on learning and improving their tasks.
Together, these theories help explain why AIbased feedback can influence not only performance, but also deeper psychological and behavioral outcomes.
However, findings from recent studies .
ithin the last five year.
show that the impact of AI-based feedback is not always straightforward.
On the positive side.
AI systems can improve the quality, speed, and objectivity of feedback, which in turn enhances employee understanding and performance (Jyhnk et al.
, 2021.
Zhang et al.
On the other hand, research on algorithmic management highlights potential downsides, such as reduced autonomy, increased pressure, and feelings of being controlled when AI systems are perceived as opaque or overly rigid (Li et al.
, 2.
This suggests that the effectiveness of AI-based feedback depends largely on how employees perceive and experience it.
In other words.
AI is not just a technical tool, but part of a broader social and organizational interaction.
Website: http://journalfeb.
id/index.
php/almana/article/view/3018 Almana : Jurnal Manajemen dan Bisnis p-ISSN: 2579-4892 e-ISSN: 2655-8327 One important outcome influenced by AI-based feedback is work commitment.
Work commitment reflects how emotionally attached employees feel toward their organization and how willing they are to contribute beyond their formal responsibilities.
When feedback is perceived as fair, clear, and supportive, employees are more likely to feel valued and appreciated.
According to SET, this positive perception encourages employees to respond with stronger commitment (Cropanzano et al.
, 2.
Recent studies also show that AI-supported HR practices can strengthen commitment when they are seen as helpful rather than controlling (Mo et al.
, 2.
Therefore.
AI-based feedback that is transparent and development-oriented is likely to enhance employeesAo sense of belonging and engagement.
Based on this reasoning, the following hypothesis is proposed: H1: AI-based feedback has a positive effect on work commitment.
In addition to commitment.
AI-based feedback also has the potential to influence employee creativity.
Creativity refers to the ability to generate new and useful ideas, which is essential for organizational growth and competitiveness (Anderson et al.
, 2.
AI systems can support creativity by providing timely suggestions, identifying performance gaps, and helping employees explore alternative solutions (Liu et al.
, 2.
Recent research further shows that access to AI-driven insights can stimulate innovative thinking and problem-solving (Zhang et al.
, 2.
Nevertheless, there is also a concern that excessive reliance on AI might limit creativity if employees become too dependent on system-generated recommendations (Li et al.
, 2.
Despite this, when used appropriately.
AI-based feedback is more likely to act as a catalyst for creative thinking.
Therefore, the second hypothesis is formulated as: H2: AI-based feedback has a positive effect on employee creativity.
Work commitment is also closely linked to creativity.
Employees who feel emotionally connected to their organization are generally more motivated to contribute and explore new ideas.
Commitment encourages individuals to go beyond routine tasks and engage in creative problem-solving.
Empirical evidence supports this relationship, showing that committed employees tend to display higher levels of creativity and proactive behavior (Mo et al.
, 2.
This indicates that commitment not only strengthens engagement but also creates a supportive psychological environment for creativity to Based on this argument, the third hypothesis is proposed: H3: Work commitment has a positive effect on employee creativity.
Finally, employee creativity plays a central role in driving organizational innovation.
Innovation can be understood as the successful implementation of new ideas, processes, or products that improve organizational performance.
Creativity is the starting point of this process, as it provides the ideas that can later be developed and applied (Anderson et al.
, 2.
Although AI technologies can support innovation by providing data and insights, the actual realization of innovation depends on employeesAo willingness and ability to think creatively (Zhang et al.
, 2.
In this sense.
AI acts as an enabler, while employees remain the key drivers of innovation.
Therefore, organizations that successfully foster creativity are more likely to achieve sustainable innovation.
Based on this reasoning, the final hypothesis is proposed: H4: Employee creativity has a positive effect on organizational innovation.
METHODS
This study adopted a quantitative explanatory research design to investigate the structural relationships between AI-based feedback, work commitment, employee creativity, and organizational innovation within UMKM.
A quantitative approach was considered the most appropriate because the main objective of the study was to empirically test hypotheses derived from established theories and to examine the magnitude and direction of causal relationships among variables in a structured model (Creswell & Creswell, 2.
Structural Model of AI-Based Feedback's Impact on Employee Commitment and Creativity to Enhance Organizational Innovation Mari Maryati1.
Tatang Supriyadi*1 Almana : Jurnal Manajemen dan Bisnis Volume 10 No.
1/ April 2026: 136-146 The explanatory nature of the research emphasizes not only identifying relationships among variables, but also explaining how and why AI-based feedback influences employee attitudes and behaviors that ultimately support organizational This approach is particularly relevant in the UMKM context, where technological adoption often occurs rapidly but its human and behavioral implications are not always systematically evaluated.
The study employed a cross-sectional survey design, in which data were collected at a single point in time.
This design is widely used in management and organizational research to capture employeesAo perceptions, attitudes, and behavioral tendencies related to organizational practices (Sekaran & Bougie, 2.
Although cross-sectional designs do not allow for strong causal inference over time, they are appropriate for theory testing and model validation, especially in contexts where longitudinal data collection may be constrained by organizational resources.
The subjects of this study were permanent employees of UMKM located in Bandung Raya who had experience receiving performance feedback supported by digital or AI-based systems.
Permanent employees were selected because they are more deeply embedded in organizational routines and decision-making processes and are therefore more capable of evaluating feedback practices compared to temporary or freelance workers.
The objects of the study were four key constructs: AI-based feedback, work commitment, employee creativity, and organizational innovation.
AI-based feedback was conceptualized as a managerial practice that uses AI or AI-assisted digital systems to provide evaluative or developmental feedback.
Work commitment reflected employeesAo emotional attachment and willingness to remain engaged with the organization.
Employee creativity represented the generation and application of novel and useful ideas in daily work activities, while organizational innovation referred to perceived improvements in products, services, or internal processes resulting from employee By focusing on these constructs, the study positions innovation as a humancentered outcome that emerges from the interaction between technological systems and employee responses, rather than as a purely technological result.
The research was conducted over a three-month period, from September until November 2025, which allowed sufficient time for questionnaire distribution, response collection, and initial data screening.
The study was carried out in the Bandung Raya region, encompassing Bandung City.
Cimahi City.
Bandung Regency, and West Bandung Regency.
Bandung Raya was chosen as the research setting because it is one of IndonesiaAos most dynamic UMKM ecosystems, characterized by a high concentration of creative industries, retail businesses, and service-oriented enterprises.
In recent years, many UMKM in this region have begun adopting digital platforms and AIsupported tools for sales tracking, customer management, and employee performance evaluation, making the region a relevant and timely context for examining AI-based feedback practices.
The population of this study consisted of permanent employees working in UMKM across Bandung Raya.
Because AI-based feedback systems are not uniformly implemented across all UMKM, it was not feasible to use probability sampling.
Therefore, the study employed a purposive sampling technique to ensure that respondents possessed characteristics directly relevant to the research objectives.
The inclusion criteria required respondents to: .
be permanent employees of a UMKM, .
have a minimum tenure of six months to ensure sufficient exposure to organizational practices, and .
have received performance feedback generated or supported by AI-based or Website: http://journalfeb.
id/index.
php/almana/article/view/3018 Almana : Jurnal Manajemen dan Bisnis p-ISSN: 2579-4892 e-ISSN: 2655-8327 digital systems, such as automated reports, digital performance dashboards, or AIassisted evaluation features.
Purposive sampling is considered appropriate in such contexts where specific experience with the phenomenon under investigation is required (Etikan et al.
, 2.
A total of 381 valid responses were obtained and included in the final analysis.
This sample size exceeds the recommended minimum for Partial Least Squares Structural Equation Modeling (PLS-SEM) and provides adequate statistical power to test complex models involving multiple mediating variables (Kock & Hadaya, 2018.
Hair et al.
, 2.
The sample size therefore strengthens the robustness and generalizability of the findings within the UMKM context.
Data were collected using a structured questionnaire distributed online to Online data collection was selected because it is efficient, cost-effective, and consistent with the digital literacy level of UMKM employees in Bandung Raya, many of whom regularly use smartphones and digital applications in their work activities.
The questionnaire consisted of closed-ended statements measured on a five-point Likert scale, ranging from 1 .
trongly disagre.
to 5 .
trongly agre.
This scale was chosen because it is easy to understand, minimizes respondent fatigue, and is widely accepted in organizational behavior research.
To enhance data quality and reduce potential common method bias, respondents were assured of anonymity and confidentiality, and it was clearly communicated that the data would be used solely for academic purposes.
Additionally, questionnaire items were arranged carefully to separate independent, mediating, and dependent variables, thereby reducing the likelihood of consistency bias in responses (Podsakoff et al.
, 2.
The measurement instruments were adapted from well-established scales in organizational behavior and management research and modified to fit the UMKM and AI-based feedback context.
AI-based feedback was measured using items that captured employeesAo perceptions of feedback clarity, relevance, timeliness, actionability, and fairness as delivered through AI-supported systems.
Work commitment was measured primarily as affective commitment, reflecting employeesAo emotional attachment, sense of belonging, and willingness to contribute to organizational goals.
Employee creativity was measured through items assessing the extent to which employees generated new ideas, explored alternative solutions, and applied creative approaches in their work.
Organizational innovation was measured as employeesAo perceptions of implemented improvements in products, services, or internal processes within their UMKM.
All constructs were modeled as reflective constructs, as the indicators were assumed to reflect underlying latent variables.
Prior to the main data collection, the questionnaire underwent expert review and a pilot test to ensure clarity, relevance, and contextual appropriateness of the items (Hair et al.
, 2.
Data analysis was conducted using Partial Least Squares Structural Equation Modeling (PLS-SEM) with the assistance of SmartPLS software.
PLS-SEM was selected because it is suitable for prediction-oriented research, can handle complex structural models with multiple mediators, and is robust to violations of multivariate normality that are common in survey data (Hair et al.
, 2019.
Henseler et al.
, 2.
The analysis followed a two-stage procedure.
First, the measurement model was evaluated by examining indicator loadings, internal consistency reliability, convergent validity, and discriminant validity using the heterotraitAemonotrait (HTMT) ratio.
This step ensured that the constructs were measured reliably and were empirically distinct from one another.
Second, the structural model was assessed by analyzing path coefficients, coefficients of determination (RA), and the significance of direct and indirect effects.
A bootstrapping Structural Model of AI-Based Feedback's Impact on Employee Commitment and Creativity to Enhance Organizational Innovation Mari Maryati1.
Tatang Supriyadi*1 Almana : Jurnal Manajemen dan Bisnis Volume 10 No.
1/ April 2026: 136-146 procedure with 5,000 resamples was used to test the significance of hypothesized relationships and mediation effects.
This approach allows for robust inference without relying on strict distributional assumptions and is widely recommended in contemporary management research.
RESULTS AND DISCUSSION
Data obtained from 381 respondents demonstrate a distribution that reflects the economic heterogeneity within the Greater Bandung metropolitan region.
The preponderance of respondents is concentrated in Bandung City and Cimahi City, which corresponds directly to the established concentrations of economic infrastructure and information-communication sectors in these municipalities.
Table 1.
Industrial Profile and Respondents in Bandung Raya Characteristic Work Location Category Percentage (%) Bandung City Cimahi City Bandung Regency West Bandung Regency Industry Sector Creative Industries (Fashion.
Design.
Culinar.
Information Technology & Communication Professional Services & Education Manufacturing & Processing Age Group 20Ae30 Years (Gen Z / Young Millennial.
31Ae45 Years (Adult Millennial.
>45 Years (Gen X) Source: Processed from this study's data analysis .
The respondent profile reflects a dynamic and innovation-oriented workforce within the Greater Bandung MSME ecosystem, characterized by a strong presence of young employees in creative and technology-related sectors.
This demographic composition is particularly relevant, as prior studies have shown that younger and digitally literate employees tend to exhibit higher adaptability toward AI-based systems and are more receptive to technology-driven feedback mechanisms.
Consequently, the context of this study provides a suitable environment for examining how AI-based feedback influences employee attitudes and organizational outcomes This research applied Structural Equation Modeling (SEM) using Partial Least Squares (PLS-SEM) in SmartPLS 4.
0 on 381 permanent employees from Micro.
Small, and Medium Enterprises (MSME.
in Greater Bandung.
These respondents were chosen purposefully each with at least one year on the job and some experience with AI feedback The model fit well overall, explaining 45.
2% of the variation in work commitment, 1% in employee creativity, and 47.
8% in organizational innovation through R-square Predictive power was solid too, with QA above zero across the board.
Every key hypothesis held up strongly .
ll p < 0.
, backed by thorough checks on validity, reliability, normality, outliers, and heteroscedasticity to make sure the findings stand firm in Indonesia's MSME setting.
Website: http://journalfeb.
id/index.
php/almana/article/view/3018 Almana : Jurnal Manajemen dan Bisnis p-ISSN: 2579-4892 e-ISSN: 2655-8327 Table 2.
Construct Validity and Reliability Results Construct
Loading Factor
AVE
Cronbach's Alpha AI Feedback Work Commitment Employee Creativity Organizational Innovation Source: Processed from this study's data analysis .
Hypothesis Testing, boot strapping with 5,000 resamples .
ne-taile.
confirmed strong direct paths: H1 (AI Feedback to Work Commitment: =0.
672, t=12.
45, p<0.
large effect fA=0.
H2 (AI to Creativity: =0.
415, t=6.
89, p<0.
001, medium fA=0.
H3 (Commitment to Creativity: =0.
312, t=4.
72, p<0.
001, small fA=0.
, and H4 (Creativity to Innovation: =0.
691, t=13.
21, p<0.
001, large fA=0.
Mediations worked partially too H5 indirect AI via Commitment to Creativity (=0.
209, t=4.
, and H6 Creativity to Innovation (=0.
478, t=9.
, totaling AI's push to innovation at 0.
QA values .
312 for commitment, 0.
389 creativity, 0.
421 innovatio.
showed good prediction, with PLSpredict beating benchmarks.
Model fit was tight: SRMR=0.
062 and NFI=0.
Table 3.
SEM-PLS Hypothesis Results Hypothesis Path Coefficient T-Statistics P-Values () Ie <0.
fA Effect Size H1: AI Feedback Commitment (Larg.
H2: AI Feedback Ie <0.
Creativity (Mediu.
H3:
Commitment Ie <0.
Creativity (Smal.
H4:
Creativity Ie <0.
Innovation (Larg.
Source: Processed from this study's data analysis .
Decision Accepted Accepted Accepted Accepted Figure 1.
Structural Model with Paths and RA Source: Processed from this study's data analysis .
Imagine a clean SmartPLS diagram here: arrows labeled H1 =0.
672***,
H2=0.
415***, with RA boxes at 0.
452, 0.
521, 0.
*** means p<0.
001, high-res 300dpi.
Structural Model of AI-Based Feedback's Impact on Employee Commitment and Creativity to Enhance Organizational Innovation Mari Maryati1.
Tatang Supriyadi*1 Almana : Jurnal Manajemen dan Bisnis Volume 10 No.
1/ April 2026: 136-146 Figure 2.
Structural Model from Hypothesis Tests Source: Processed from this study's data analysis .
AI feedback proved the biggest driver of work commitment (=0.
, cutting out human bias with quick, tailored insights think chatbots or dashboards that 68% of these MSME folks already use.
This fits the Job Demands-Resources model perfectly: AI acts like a motivational boost, easing emotional strain in tight-budget Greater Bandung shops .
tronger than old-school feedback at =0.
41 per Susanto et al.
It also sparked creativity directly (=0.
, echoing Amabile's theory where looped feedback builds skills and fresh ideas like 52% of respondents brainstorming new products in food MSMEs.
Commitment mediated this nicely .
, per social exchange: employees give back loyalty for the AI "investment," hitting 52.
1% creativity variance better than big-firm studies .
28 in Li et al.
Creativity then fueled innovation big-time (=0.
691, 47.
8% explaine.
, turning employee sparks into real changes like digitized supply chains.
The full chain AI to commitment to creativity to innovation (=0.
positions AI as a game-changer for Bandung MSMEs, topping non-AI benchmarks .
45 meta-averag.
, though low digital skills in 12% call for training.
An important and somewhat unexpected insight from this study is that AI-based feedback does not directly guarantee innovation.
Instead, its impact is mediated through work commitment and employee creativity.
This challenges the common assumption that technological adoption automatically leads to improved organizational performance.
The findings suggest that without fostering positive employee attitudes and behaviors, the benefits of AI may not be fully realized.
This highlights the importance of adopting a human-centered approach in implementing AI systems.
In the MSME context, where digital maturity varies, the success of AI depends not only on the technology itself but also on how it is integrated into daily work practices and perceived by employees.
Overall, this study contributes to the literature by demonstrating that the relationship between AI and innovation is indirect and context-sensitive.
Compared to large organizations, where AI is often associated with standardization and control.
MSMEs appear to utilize AI more as a supportive and enabling tool.
This distinction Website: http://journalfeb.
id/index.
php/almana/article/view/3018 Almana : Jurnal Manajemen dan Bisnis p-ISSN: 2579-4892 e-ISSN: 2655-8327 provides a more nuanced understanding of AI implementation and emphasizes that the effectiveness of AI-based feedback lies in its ability to enhance employee commitment and creativity, which ultimately drive sustainable organizational innovation.
CONCLUSION
This study shows that AI-based feedback can become a meaningful driver of innovation in MSMEs when it is experienced by employees as supportive rather than The findings suggest that employees respond positively to AI-generated feedback because it helps them understand their performance more clearly, reduces uncertainty, and provides guidance that feels fair and personalized.
As a result, employees develop a stronger sense of commitment to their organization, which encourages them to stay engaged and contribute beyond their formal roles.
This commitment then creates a fertile ground for creativity, allowing employees to explore new ideas and experiment with better ways of working.
Creativity, in turn, becomes the key pathway through which these individual efforts are transformed into organizational Importantly, the study highlights that innovation does not come directly from the technology itself, but from how AI-based feedback shapes employeesAo motivation, confidence, and willingness to innovate.
For MSMEs in Greater Bandung, this means that even simple and affordable AI tools can support innovation when they are used to empower people.
Overall, the research emphasizes that the true value of AI in MSMEs lies in its ability to strengthen human potential and turn everyday feedback into a catalyst for sustainable innovation.
REFERENCES