In e na ional Jou nal o Eme ging Resea ch in Science, Enginee ing, and Managemen
Vol. 1, Issue 4, pp.09-18, Oc obe 2025.
www.ije sem.com eISSN – 3107-9075
IJERSEM@2025 h ps://doi.o g/10.58482/ije sem. 1i4.2 9
Impac o A i icial In elligence Adop ion on
Employee P oduc i i y in SMEs
A. Sahana
Associa e P o esso , Mas e o Business Adminis a ion, The Ox o d College o Enginee ing, Bengulu u, India
Abs ac : The adop ion o A i icial In elligence (AI) echnologies is inc easingly ans o ming he ope a ions o small and medium-sized
en e p ises (SMEs), in luencing employee p oduc i i y (EP) h ough au oma ion, decision suppo , and adap i e lea ning sys ems. This pape
in es iga es he mul i ace ed e ec s o AI adop ion on employee p oduc i i y in SMEs, d awing insigh s om ecen empi ical and heo e ical
s udies. The analysis explo es bo h enable s and inhibi o s o p oduc i i y, emphasizing he media ing oles o o ganiza ional cul u e, employee
au onomy, and echnos ess managemen . S udies in he li e a u e sugges ha AI in eg a ion can signi ican ly enhance EP when combined wi h
e hical go e nance and suppo i e wo k en i onmen s. Con e sely, un egula ed AI deploymen may induce objec i ica ion, job anxie y, and
educed engagemen . To add ess hese challenges, his pape p oposes a concep ual amewo k linking AI-enabled p ocess inno a ion, ask
edesign, and upskilling o sus ained p oduc i i y gains. The indings highligh ha SMEs should s a egically align AI adop ion wi h human-
cen ic p ac ices o ensu e ha p oduc i i y imp o emen s a e bo h measu able and sus ainable.
Keywo ds: A i icial In elligence, Employee P oduc i i y, Small and Medium-sized En e p ises, O ganiza ional Pe o mance, Wo kplace
Inno a ion.
1 INTRODUCTION
A i icial In elligence (AI) has eme ged as a ans o ma i e o ce eshaping he global business landscape, o e ing
o ganiza ions new oppo uni ies o enhance ope a ional e iciency, inno a ion, and decision-making quali y. Fo small and
medium-sized en e p ises (SMEs), which cons i u e mo e han 90% o global businesses and con ibu e signi ican ly o
employmen gene a ion, AI adop ion has he po en ial o unlock new le els o employee p oduc i i y (EP) and compe i i eness
[1]. Howe e , he in eg a ion o AI in o o ganiza ional wo k lows b ings bo h oppo uni ies and challenges ha mus be unde s ood
in a human-cen e ed con ex .
Employee p oduc i i y in SMEs is in luenced no only by echnology adop ion bu also by cul u al, manage ial, and
psychological ac o s [2]. While AI sys ems can au oma e ou ine asks, op imize wo k lows, and assis in da a-d i en decision-
making, hei impac on human pe o mance depends on how employees pe cei e and in e ac wi h hese sys ems. S udies indica e
ha when AI ools complemen a he han eplace human labo , hey can signi ican ly enhance pe o mance, educe cogni i e
load, and imp o e c ea i i y and decision speed [3], [4]. Con e sely, excessi e eliance on AI au oma ion wi hou su icien human
con ol can lead o employee disengagemen , echnos ess, and diminished us [5].
The le el o AI eadiness among SMEs emains une en ac oss indus ies and egions. Resea ch by Gup a e al. [2] highligh s
ha SMEs o en ace esou ce limi a ions, lack o echnical expe ise, and insu icien digi al in as uc u e, which impede e ec i e
AI in eg a ion. Kassa and Wo ku [1] a gue ha leade ship suppo and o ganiza ional lea ning play c ucial oles in ealizing
p oduc i i y gains om digi al ans o ma ion. Mo eo e , Chen e al. [3] emphasize ha aligning AI ools wi h employees’ ask
equi emen s and au onomy can p oduce measu able imp o emen s in ou pu quali y and e iciency.
F om a beha io al pe spec i e, he in e ac ion be ween human employees and AI sys ems is shaped by pe cei ed use ulness,
pe cei ed h ea , and o ganiza ional suppo mechanisms [6]. Employees who eel empowe ed o use AI ools e ec i ely end o
epo highe job sa is ac ion and p oduc i i y, while hose who ea eplacemen o su eillance exhibi esis ance and s ess. This
duali y unde sco es he impo ance o adop ing a human–AI collabo a ion model, whe e AI is posi ioned as an augmen a i e
pa ne a he han a subs i u e o human capabili y [7], [8].
AI’s impac on p oduc i i y is also media ed by o ganiza ional p ocesses such as ask edesign, wo k low in eg a ion, and
con inuous lea ning. Schola s no e ha SMEs adop ing AI o p ocess op imiza ion expe ience as e u na ound imes and
imp o ed ou pu consis ency, i aining and knowledge-sha ing mechanisms a e in place [9], [10]. E hical and go e nance aspec s
u he in luence p oduc i i y ou comes. T anspa en AI policies and esponsible da a p ac ices os e employee con idence and
educe anxie y ela ed o ai ness and job secu i y [11]. Conside ing hese indings, his pape examines he dynamic ela ionship
be ween AI adop ion and employee p oduc i i y in SMEs. I seeks o add ess he ollowing esea ch objec i es:
1. To iden i y he key enable s and ba ie s in luencing AI-d i en p oduc i i y in SMEs.
2. To analyze how o ganiza ional cul u e, leade ship, and employee a i udes media e AI’s p oduc i i y ou comes.
In e na ional Jou nal o Eme ging Resea ch in Science, Enginee ing, and Managemen
Vol. 1, Issue 4, pp.09-18, Oc obe 2025.
www.ije sem.com eISSN – 3107-9075
IJERSEM@2025 h ps://doi.o g/10.58482/ije sem. 1i4.2 10
3. To de elop a concep ual amewo k linking AI adop ion s a egies o sus ained employee p oduc i i y and o ganiza ional
g ow h.
The emainde o his pape is o ganized as ollows: Sec ion 2 e iews he exis ing li e a u e on AI adop ion and employee
p oduc i i y. Sec ion 3 p esen s he p oposed concep ual amewo k. Sec ion 4 ou lines he esea ch me hodology, while Sec ion
5 discusses esul s and implica ions. Sec ion 6 concludes wi h ecommenda ions o SME manage s and policymake s.
2 LITERATURE REVIEW
A i icial In elligence (AI) adop ion in small and medium-sized en e p ises (SMEs) has become a de ining ac o o
o ganiza ional compe i i eness in he digi al e a. P io esea ch b oadly ca ego izes he impac o AI adop ion on employee
p oduc i i y (EP) in o h ee domains: (i) p oduc i i y enable s, (ii) p oduc i i y inhibi o s, and (iii) media ing o mode a ing ac o s
ha de e mine he s eng h and di ec ion o AI’s in luence on human pe o mance.
2.1 AI Adop ion as a P oduc i i y Enable
AI enhances p oduc i i y p ima ily by au oma ing epe i i e asks, augmen ing decision-making, and acili a ing inno a ion.
S udies e eal ha AI in eg a ion educes ope a ional ine iciencies and imp o es employee pe o mance when aligned wi h
s a egic goals and human esou ce de elopmen [1], [2]. Gup a e al. [2] obse ed ha AI adop ion imp o es decision quali y,
esponsi eness, and ou pu accu acy h ough da a-d i en insigh s. Simila ly, Chen e al. [3] demons a ed ha AI-augmen ed
decision sys ems in manu ac u ing SMEs inc eased h oughpu and ask p ecision while educing manual e o a es.
AI-d i en ools such as in elligen p ocess au oma ion, p edic i e analy ics, and na u al language p ocessing ha e been shown
o signi ican ly educe employees’ cogni i e load, allowing hem o ocus on high- alue c ea i e and analy ical asks [4]. Ebe ha d
[7] u he asse s ha AI unc ions as a pe o mance enhance when iewed as a collabo a i e augmen a ion a he han a
eplacemen o human in elligence. This pe spec i e aligns wi h Das and Sheikh [8], who highligh ha human-cen ic AI adop ion
imp o es psychological well-being, job sa is ac ion, and pe o mance sus ainabili y.
The p oduc i i y-enabling e ec s o AI a e mos e iden in SMEs ha in eg a e AI-enabled p ocess eenginee ing and
con inuous upskilling p og ams [9], [10]. Kama a [9] ound ha AI deploymen in wo k low au oma ion imp o ed p oduc quali y
consis ency, while Thomas and C uz [10] emphasized ha digi ally li e a e employees adap mo e apidly o AI-d i en sys ems,
hus main aining compe i i e p oduc i i y le els. These indings collec i ely indica e ha AI’s ole as a p oduc i i y d i e
depends hea ily on alignmen wi h o ganiza ional lea ning and inno a ion s a egies.
2.2 AI Adop ion as a P oduc i i y Inhibi o
While AI can be ans o ma i e, i s adop ion can also gene a e ad e se ou comes when inadequa ely managed. Technos ess,
job insecu i y, and e hical ambigui y a e among he mos ci ed inhibi o s o employee p oduc i i y in AI-in eg a ed wo kplaces
[5], [6]. Rich e and Unge [5] ound ha employees exposed o high le els o au oma ion anxie y and su eillance epo ed
dec eased job sa is ac ion and educed mo i a ion. The pe cep ion o being cons an ly e alua ed by AI-based moni o ing sys ems
can e ode us and engagemen , leading o pe o mance de e io a ion.
Huang e al. [6] iden i ied psychological adap a ion challenges among SME employees ansi ioning o AI-suppo ed
en i onmen s, whe e unce ain y abou AI’s decision logic os e s esis ance. Simila indings by Annamalai e al. [11] show ha
he absence o e hical anspa ency and clea communica ion in AI implemen a ion leads o mis us and mo al a igue, nega i ely
a ec ing pe o mance. Fu he mo e, SMEs wi h limi ed echnical capaci y o en s uggle o p o ide adequa e aining o change
managemen suppo , compounding s ess and p oduc i i y loss.
Ano he inhibi o y ac o is skill obsolescence. As AI au oma es knowledge-in ensi e asks, employees lacking digi al
compe encies ace ole edundancy o educed ask signi icance [1]. This misalignmen be ween human capabili ies and AI
unc ionali ies con ibu es o pe cei ed inequi y and diminished in insic mo i a ion. Consequen ly, AI’s p oduc i i y bene i s
canno be gene alized wi hou conside ing con ex -speci ic o ganiza ional and human ac o s.
2.3 Media ing and Mode a ing Fac o s In luencing P oduc i i y Ou comes
A g owing body o li e a u e emphasizes ha AI’s impac on p oduc i i y is no di ec , bu media ed by o ganiza ional,
indi idual, and echnological a iables [2], [3], [8]. Among hese, o ganiza ional cul u e and leade ship o ien a ion a e key
de e minan s. Kassa and Wo ku [1] obse ed ha ans o ma ional leade ship os e s us and collabo a ion in AI adop ion,
enabling employees o pe cei e echnology as an enable a he han a h ea . Con e sely, igid hie a chical cul u es hinde
adap a ion and exace ba e esis ance.
In e na ional Jou nal o Eme ging Resea ch in Science, Enginee ing, and Managemen
Vol. 1, Issue 4, pp.09-18, Oc obe 2025.
www.ije sem.com eISSN – 3107-9075
IJERSEM@2025 h ps://doi.o g/10.58482/ije sem. 1i4.2 11
Employee au onomy also mode a es he AI-p oduc i i y ela ionship. Chen e al. [3] showed ha when employees e ain
decision-making con ol in AI-augmen ed en i onmen s, hei p oduc i i y le els ise signi ican ly. Simila ly, Ebe ha d [7] ound
ha human–AI collabo a ion h i es unde decen alized s uc u es ha encou age expe imen a ion and sha ed accoun abili y.
Technological eadiness ac s as bo h a p e equisi e and a mode a ing condi ion. Gup a e al. [2] highligh ed ha AI eadiness—
de ined by digi al in as uc u e, da a go e nance, and skill a ailabili y—di ec ly in luences p oduc i i y ou comes. SMEs wi h
low eadiness expe ience delayed implemen a ion and ine icien u iliza ion o AI ools. E hical go e nance, as discussed by
Annamalai e al. [11], u he media es his ela ionship by enhancing anspa ency, ensu ing da a p o ec ion, and imp o ing
employees’ con idence in AI sys ems.
Con inuous lea ning ecosys ems ep esen ano he c i ical media o . Thomas and C uz [10] demons a ed ha SMEs
inco po a ing digi al lea ning p og ams in o hei AI s a egies expe ienced s eady p oduc i i y imp o emen s due o enhanced
adap abili y and echnical compe ence. Das and Sheikh [8] suppo his iew, emphasizing ha aining and e hical o ien a ion
mi iga e echnos ess while ein o cing human–AI us .
2.4 Syn hesis o he Li e a u e
The e iewed li e a u e indica es ha AI’s impac on employee p oduc i i y is mul idimensional, shaped by bo h echnical and
human ac o s. When in eg a ed esponsibly—suppo ed by e hical go e nance, anspa en communica ion, and skill
de elopmen —AI ac s as a ca alys o p oduc i i y. Howe e , unmanaged AI adop ion isks inducing job anxie y, echnos ess,
and esis ance, especially in esou ce-cons ained SMEs. The in e play o AI capabili y, o ganiza ional cul u e, leade ship s yle,
and employee adap abili y de e mines whe he AI ul ima ely enhances o diminishes p oduc i i y. This syn hesis unde sco es he
need o a holis ic concep ual amewo k ha in eg a es echnological, human, and o ganiza ional pe spec i es o explain how AI
adop ion in luences employee p oduc i i y. The nex sec ion (Sec ion 3) p esen s such a amewo k, designed o cap u e he
dynamic in e ac ions among hese ac o s wi hin he SME con ex .
3 CONCEPTUAL FRAMEWORK
The e iewed li e a u e es ablishes ha he ela ionship be ween A i icial In elligence (AI) adop ion and employee
p oduc i i y (EP) in small and medium-sized en e p ises (SMEs) is complex, mul idi ec ional, and in luenced by se e al
o ganiza ional and indi idual-le el a iables. This sec ion p oposes a concep ual amewo k ha syn hesizes hese a iables in o
an in eg a ed model explaining how AI adop ion a ec s employee p oduc i i y h ough media ing and mode a ing mechanisms.
3.1 F amewo k O e iew
The p oposed amewo k (Fig. 1) is buil upon h ee ounda ional p emises:
1. AI as a Capabili y D i e : AI echnologies enhance employee p oduc i i y by au oma ing ou ine p ocesses, imp o ing
decision quali y, and enabling in elligen suppo sys ems [2], [3], [4].
2. Human–O ganiza ional Media ion: The e ec i eness o AI depends on how employees in e ac wi h hese sys ems,
shaped by o ganiza ional cul u e, leade ship, au onomy, and e hical en i onmen [1], [7], [8], [11].
3. Feedback and Adap a ion: P oduc i i y ou comes u he in luence u u e AI adop ion and digi al lea ning, o ming a
eedback loop ha s eng hens o ganiza ional adap abili y and compe i i eness [9], [10].
3.2 Componen s o he F amewo k
(a) AI Adop ion
AI adop ion e e s o he in eg a ion o in elligen echnologies—such as p edic i e analy ics, machine lea ning, and p ocess
au oma ion—in o co e SME ope a ions. The ex en and ma u i y o AI adop ion de e mine i s po en ial o d i e inno a ion and
pe o mance [2]. SMEs wi h high AI eadiness and adequa e digi al in as uc u e end o expe ience g ea e p oduc i i y
imp o emen s han hose wi h limi ed esou ces [1].
(b) Media ing Va iables
Se e al media ing ac o s b idge he link be ween AI adop ion and employee p oduc i i y:
1. O ganiza ional Cul u e and Leade ship: T ans o ma ional and pa icipa i e leade ship s yles c ea e us and openness
owa d AI echnologies. Suppo i e cul u es encou age expe imen a ion, educe ea o ailu e, and p omo e AI-enabled
lea ning [1], [7].
2. Employee Au onomy and Skill De elopmen : When employees main ain con ol o e AI-assis ed decisions and ha e
access o con inuous lea ning oppo uni ies, AI’s bene i s ampli y [3], [8], [10]. Au onomy os e s owne ship,
accoun abili y, and c ea i e engagemen , leading o sus ained pe o mance imp o emen s.
In e na ional Jou nal o Eme ging Resea ch in Science, Enginee ing, and Managemen
Vol. 1, Issue 4, pp.09-18, Oc obe 2025.
www.ije sem.com eISSN – 3107-9075
IJERSEM@2025 h ps://doi.o g/10.58482/ije sem. 1i4.2 12
3. Technos ess Managemen : Excessi e au oma ion and AI-d i en su eillance may cause anxie y, o e load, o esis ance
[5], [6]. E ec i e s ess mi iga ion— h ough anspa en communica ion, e gonomic digi al ools, and men al heal h
suppo —ensu es a smoo he adop ion p ocess and s able p oduc i i y le els.
4. E hical and Go e nance P ac ices: E hical anspa ency, ai ness in AI decisions, and p i acy p o ec ion inc ease
employees’ us and willingness o engage wi h in elligen sys ems [11]. These ac o s educe cogni i e dissonance and
imp o e o e all pe o mance sa is ac ion.
(c) Employee P oduc i i y (EP)
Employee p oduc i i y ep esen s he measu able ou pu o e iciency achie ed pe employee ela i e o e o , ime, o
esou ces. AI adop ion in luences EP h ough p ocess op imiza ion, edundancy educ ion, and suppo o analy ical o c ea i e
asks [2], [4], [9]. Howe e , p oduc i i y gains a e con ingen on success ul media ion by cul u al and indi idual ac o s.
(d) Feedback Mechanism
The model inco po a es a eedback loop whe e enhanced p oduc i i y and employee adap abili y s eng hen AI capabili y
u iliza ion. Success ul implemen a ion leads o g ea e o ganiza ional lea ning and eadiness, which in u n ein o ces u u e AI
adop ion [10]. Con e sely, nega i e expe iences—such as high echnos ess o e hical con lic s—may educe willingness o use
AI ools, weakening he cycle.
3.3 Theo e ical Unde pinning
This amewo k d aws upon h ee heo e ical pe spec i es:
1. Socio-Technical Sys ems Theo y (STS): This heo y emphasizes he join op imiza ion o social and echnical subsys ems
wi hin o ganiza ions. AI mus be in eg a ed no me ely as a echnical inno a ion bu as pa o a socio-o ganiza ional
ecosys em ha alues employee inpu and adap abili y [6].
2. Job Demands–Resou ces (JD-R) Model: AI educes job demands (e.g., epe i i e asks) while p o iding esou ces (e.g.,
decision suppo ). Howe e , i AI inc eases cogni i e load o s ess, i s bene i s may be o se . P ope esou ce balancing
is he e o e c ucial [5], [8].
3. Human–AI Collabo a ion Theo y: As p oposed by Ebe ha d [7] and Das and Sheikh [8], AI should augmen human
capabili ies, no eplace hem. The syne gy be ween algo i hmic in elligence and human judgmen leads o supe io
p oduc i i y ou comes.
3.4 Concep ual Model Desc ip ion
The concep ual amewo k shown in Fig. 1 posi s ha :
• AI Adop ion di ec ly in luences Employee P oduc i i y.
• The s eng h o his ela ionship is media ed by O ganiza ional Cul u e, Employee Au onomy, Technos ess Managemen ,
and E hical Go e nance.
• Con inuous Lea ning and Feedback Mechanisms sus ain p oduc i i y gains by imp o ing AI u iliza ion and eadiness.
• E hical and Human-cen ic Implemen a ion ac s as a s abilizing ac o , ensu ing ha AI adop ion emains p oduc i e and
psychologically sus ainable.
Fig. 1. Concep ual F amewo k o AI adop ion and Employee P oduc i i y in SMEs
In e na ional Jou nal o Eme ging Resea ch in Science, Enginee ing, and Managemen
Vol. 1, Issue 4, pp.09-18, Oc obe 2025.
www.ije sem.com eISSN – 3107-9075
IJERSEM@2025 h ps://doi.o g/10.58482/ije sem. 1i4.2 13
4 RESEARCH METHODOLOGY
4.1 Resea ch Design
This s udy adop s a quan i a i e, c oss-sec ional esea ch design aimed a empi ically alida ing he concep ual ela ionships
p oposed in Sec ion 3. The esea ch in es iga es how AI adop ion in luences employee p oduc i i y in small and medium-sized
en e p ises (SMEs) h ough media ing o ganiza ional and beha io al ac o s. This design aligns wi h p io quan i a i e analyses in
echnology adop ion and p oduc i i y s udies [1], [2], [3], allowing he collec ion o s anda dized esponses o s a is ical modeling
and hypo hesis es ing. A s uc u ed ques ionnai e su ey was de eloped based on es ablished cons uc s om p io s udies on AI
eadiness [2], echnos ess [5], human–AI collabo a ion [7], and e hical go e nance [11]. The su ey i ems we e measu ed on a
i e-poin Like scale anging om “s ongly disag ee (1)” o “s ongly ag ee (5).”
4.2 Popula ion and Sampling
The a ge popula ion comp ised employees and middle-le el manage s wo king in AI-adop ing SMEs ac oss he
manu ac u ing, se ice, and IT sec o s. SMEs we e selec ed acco ding o he de ini ions o he Minis y o Mic o, Small and
Medium En e p ises (MSME), India—o ganiza ions wi h ewe han 250 employees and u no e below ₹250 c o e. A s a i ied
andom sampling echnique was applied o ensu e sec o al ep esen a ion. In i a ions we e sen o 450 employees ac oss 60 SMEs,
ou o which 312 alid esponses we e ecei ed ( esponse a e: 69.3%). This sample size exceeds he ecommended minimum o
200 o s uc u al equa ion modeling (SEM), ensu ing adequa e s a is ical powe [3], [8].
4.3 Da a Collec ion P ocedu e
Da a we e collec ed h ough a combina ion o online su eys and in-pe son dis ibu ion. Responden s we e b ie ed abou he
s udy objec i es and assu ed o da a con iden iali y. Pa icipa ion was olun a y, and anonymi y was main ained h oughou .
The ques ionnai e included he ollowing sec ions:
• Sec ion A: Demog aphic and o ganiza ional da a (age, gende , ole, indus y ype, i m size).
• Sec ion B: Ex en and ype o AI ools adop ed (au oma ion sys ems, decision suppo , p edic i e analy ics).
• Sec ion C: Pe cep ions ela ed o media ing ac o s—o ganiza ional cul u e, au onomy, echnos ess, and e hics.
• Sec ion D: Sel - epo ed measu es o employee p oduc i i y and job sa is ac ion.
All i ems we e p e- es ed wi h en expe s om academia and indus y o ensu e cla i y and con ex ual ele ance. The pilo es
(n = 30) yielded a C onbach’s alpha o 0.86, indica ing high in e nal eliabili y.
4.4 Measu emen o Cons uc s
The measu emen scales o each cons uc we e adap ed and alida ed om p e ious esea ch:
Table 1. Measu emen scales o each cons uc
Cons uc
Sou ce
Example I em
Reliabili y
(α)
AI Adop ion
[2], [3]
“Ou o ganiza ion uses AI-based sys ems o p ocess
op imiza ion.”
0.88
O ganiza ional Cul u e &
Leade ship
[1], [7]
“Managemen encou ages expe imen a ion wi h new AI
ools.”
0.85
Employee Au onomy & Skill
De elopmen
[3], [8], [10]
“I can make decisions independen ly when using AI
sys ems.”
0.82
Technos ess Managemen
[5], [6]
“AI ools inc ease my wo kload p essu e.” ( e e se-coded)
0.84
E hical & Go e nance
P ac ices
[11]
“Ou o ganiza ion ensu es ai and anspa en use o AI
decisions.”
0.86
Employee P oduc i i y
[4], [9]
“AI echnologies ha e imp o ed my wo k ou pu and
e iciency.”
0.90
Each cons uc was modeled as a la en a iable measu ed by h ee o i e obse ed indica o s. Con i ma o y ac o analysis
(CFA) was used o assess cons uc alidi y and con e gen eliabili y.
In e na ional Jou nal o Eme ging Resea ch in Science, Enginee ing, and Managemen
Vol. 1, Issue 4, pp.09-18, Oc obe 2025.
www.ije sem.com eISSN – 3107-9075
IJERSEM@2025 h ps://doi.o g/10.58482/ije sem. 1i4.2 14
4.5 Da a Analysis Techniques
Da a analysis was pe o med using SPSS 28 and AMOS 26 so wa e. The ollowing s eps we e unde aken:
1. Desc ip i e S a is ics: To summa ize demog aphic p o iles and ini ial pe cep ions o AI adop ion.
2. Reliabili y and Validi y Tes s: C onbach’s alpha and composi e eliabili y we e assessed o in e nal consis ency; A e age
Va iance Ex ac ed (AVE) alues abo e 0.5 con i med con e gen alidi y.
3. Co ela ion Analysis: Pea son’s co ela ion coe icien s measu ed he s eng h and di ec ion o ela ionships be ween
cons uc s.
4. S uc u al Equa ion Modeling (SEM): SEM es ed he hypo hesized ela ionships in he concep ual amewo k (Fig. 1),
including di ec and media ing e ec s o o ganiza ional and beha io al ac o s.
5. Media ion Analysis: The boo s apping me hod (5,000 esamples) was employed o es indi ec e ec s wi h 95%
con idence in e als.
6. Model Fi Indices: Accep able i h esholds we e: χ²/d < 3.0, CFI > 0.90, TLI > 0.90, RMSEA < 0.08 [9].
4.6 E hical Conside a ions
The s udy adhe ed o s anda d esea ch e hics p o ocols, including in o med consen , con iden iali y, and olun a y
pa icipa ion. No pe sonal iden i ie s we e collec ed. E hical clea ance was ob ained om he a ilia ed academic ins i u ion p io
o da a collec ion. The me hodological design ensu es bo h quan i a i e igo and con ex ual ele ance in analyzing AI-d i en
p oduc i i y ou comes in SMEs. The nex sec ion (Sec ion 5) p esen s empi ical esul s and hypo hesis es ing, highligh ing di ec ,
media ing, and eedback ela ionships as concep ualized in Fig. 1.
5 RESULTS AND DISCUSSION
5.1 Desc ip i e S a is ics
Table 2 p esen s he demog aphic and o ganiza ional cha ac e is ics o he esponden s. Among he 312 alid pa icipan s, 58%
we e male and 42% emale. Nea ly hal (47%) we e employed in manu ac u ing, 32% in in o ma ion echnology, and 21% in
se ices. A majo i y (63%) held mid-le el manage ial o echnical oles, and 71% had mo e han h ee yea s o expe ience using
digi al o AI-assis ed ools. These demog aphics indica e adequa e exposu e o AI echnologies wi hin SME ope a ions.
Table 2. Responden P o ile
Va iable
Ca ego y
Gende
Male (58), Female (42)
Indus y Type
Manu ac u ing (47), IT (32), Se ices (21)
Role
Manage ial (38), Technical (25), Ope a ional (37)
AI Expe ience
< 1 yea (12), 1–3 yea s (17), > 3 yea s (71)
5.2 Reliabili y and Validi y Assessmen
All cons uc s exceeded he C onbach’s alpha h eshold o 0.7, con i ming s ong in e nal eliabili y ( ange: 0.82–0.90). The
Composi e Reliabili y (CR) alues anged om 0.84 o 0.91, while he A e age Va iance Ex ac ed (AVE) alues exceeded 0.5,
indica ing con e gen alidi y. Disc iminan alidi y was con i med as he squa e oo o each AVE exceeded in e -cons uc
co ela ions. These esul s align wi h bes -p ac ice guidelines o s uc u al modeling [9].
5.3 S uc u al Model and Fi Indices
The hypo hesized s uc u al model (Fig. 1) was es ed using S uc u al Equa ion Modeling (SEM). The esul s indica ed a
s ong model i :
Table 3. Model Fi Indices
Fi Index
Recommended Value
Ob ained Value
χ²/d
< 3.0
2.31
CFI
> 0.90
0.943
TLI
> 0.90
0.936
RMSEA
< 0.08
0.051
The model demons a es good i , alida ing he concep ual amewo k’s p edic i e capabili y o AI adop ion and p oduc i i y
ela ionships.
In e na ional Jou nal o Eme ging Resea ch in Science, Enginee ing, and Managemen
Vol. 1, Issue 4, pp.09-18, Oc obe 2025.
www.ije sem.com eISSN – 3107-9075
IJERSEM@2025 h ps://doi.o g/10.58482/ije sem. 1i4.2 15
5.4 Hypo hesis Tes ing
The hypo heses we e es ed o bo h di ec and media ing ela ionships using SEM pa h coe icien s.
Table 4. Hypo hesis Tes ing Resul s
Hypo hesis
Rela ionship
S anda dized
β
p-
alue
Resul
H1
AI Adop ion → Employee P oduc i i y
0.41
< 0.001
Suppo ed
H2
AI Adop ion → O ganiza ional Cul u e & Leade ship
0.56
< 0.001
Suppo ed
H3
AI Adop ion → Employee Au onomy & Skill De elopmen
0.48
< 0.001
Suppo ed
H4
AI Adop ion → Technos ess Managemen
–0.27
< 0.01
Suppo ed (nega i e
e ec )
H5
AI Adop ion → E hical & Go e nance P ac ices
0.52
< 0.001
Suppo ed
H6
Media ing Va iables → Employee P oduc i i y
0.59
< 0.001
Suppo ed
The esul s show ha AI Adop ion has bo h di ec and indi ec e ec s on employee p oduc i i y. The media ing a iables
collec i ely explain 64% o he a iance in p oduc i i y (R² = 0.64), con i ming hei cen al ole in shaping AI ou comes.
5.5 Media ing E ec s
Media ion analysis using he boo s apping me hod (5,000 samples) e ealed ha h ee media o s had signi ican indi ec
e ec s:
1. O ganiza ional Cul u e and leade ship: β = 0.23, 95% CI [0.14, 0.34]; p < 0.001.
– AI adop ion os e s adap i e and inno a ion- iendly cul u es ha boos mo ale and collabo a ion [1], [7].
2. Employee Au onomy and Skill De elopmen : β = 0.19, 95% CI [0.10, 0.29]; p < 0.001.
– Skill empowe men enhances pe cei ed con ol and e iciency in AI-aided asks [3], [8], [10].
3. E hical and Go e nance P ac ices: β = 0.15, 95% CI [0.07, 0.23]; p < 0.01.
– T anspa en policies inc ease us , educing anxie y owa d au oma ion [11].
Technos ess Managemen had a pa ial media ion e ec (β = –0.08, p < 0.05), indica ing ha poo digi al e gonomics and lack
o suppo can weaken AI’s posi i e impac .
5.6 Discussion
The indings ein o ce he no ion ha AI adop ion enhances employee p oduc i i y p ima ily h ough o ganiza ional and
human-cen ic media o s. The di ec e ec (β = 0.41) suppo s p io e idence ha AI echnologies imp o e decision accu acy and
p ocess e iciency [2], [3], [4]. Howe e , he media ing s eng h (β = 0.59) unde sco es he impo ance o complemen a y
manage ial and cul u al mechanisms.
O ganiza ional Cul u e and Leade ship
A cul u e ha encou ages expe imen a ion and con inuous lea ning enhances he success o AI ini ia i es [1], [7].
T ans o ma ional leade ship beha io s—such as inspi ing inno a ion and in ol ing employees in AI decision-making—a e
c i ical o sus aining mo i a ion and p oduc i i y.
Employee Au onomy and Skill De elopmen
Consis en wi h Chen e al. [3] and Thomas and C uz [10], employees who a e empowe ed o use AI ools c ea i ely exhibi
highe ou pu quali y. Au onomy also media es he psychological accep ance o AI, educing ea s o obsolescence and enhancing
adap abili y.
Technos ess and E hical Conce ns
Rich e and Unge [5] cau ion ha uncon olled au oma ion may induce s ess and job a igue. The nega i e coe icien
obse ed (β = –0.27) con i ms ha unmanaged echnos ess can pa ially o se p oduc i i y gains. Simila ly, e hical and
go e nance ac o s [11]—such as ai ness in AI ecommenda ions and p i acy assu ance—eme ged as c ucial de e minan s o
sus ained p oduc i i y.
Feedback Mechanism
Empi ical da a also suppo he eedback assump ion in he concep ual model (Fig. 1). SMEs epo ing high p oduc i i y le els
we e mo e likely o ein es in AI ools, employee aining, and e hical da a policies, hus s eng hening o ganiza ional eadiness
o subsequen digi al ans o ma ions [8], [9], [10].
In e na ional Jou nal o Eme ging Resea ch in Science, Enginee ing, and Managemen
Vol. 1, Issue 4, pp.09-18, Oc obe 2025.
www.ije sem.com eISSN – 3107-9075
IJERSEM@2025 h ps://doi.o g/10.58482/ije sem. 1i4.2 16
5.7 Theo e ical and Manage ial Implica ions
Theo e ical Implica ions
This s udy con ibu es o he e ol ing li e a u e by empi ically alida ing a media ing amewo k linking AI adop ion o
employee p oduc i i y. The esul s ex end socio- echnical and human–AI collabo a ion heo ies [6], [7], [8] by con i ming ha
balanced in e ac ion be ween echnology and human agency leads o op imal ou comes.
Manage ial Implica ions
Fo SME manage s, he esul s emphasize ha p oduc i i y imp o emen s depend no me ely on echnology in es men bu on
os e ing us , e hical o e sigh , and lea ning-o ien ed leade ship. Policies should p io i ize:
• Ongoing AI li e acy and aining p og ams.
• T anspa en AI decision logic communica ion.
• Flexible, au onomy-suppo i e wo k s uc u es.
• Mechanisms o moni o and mi iga e echnos ess.
The s udy p o ides empi ical con i ma ion ha AI adop ion signi ican ly imp o es employee p oduc i i y in SMEs, p o ided
ha adop ion is accompanied by s ong leade ship, e hical go e nance, and employee empowe men . The nex sec ion (Sec ion 6)
concludes he s udy and p o ides ac ionable ecommenda ions and u u e esea ch di ec ions.
6 CONCLUSIONS AND RECOMMENDATIONS
6.1 Conclusions
A i icial In elligence (AI) has eme ged as a powe ul ca alys o o ganiza ional ans o ma ion, enabling SMEs o enhance
p oduc i i y, inno a ion, and compe i i eness. This s udy in es iga ed he impac o AI adop ion on employee p oduc i i y (EP)
wi hin small and medium-sized en e p ises, ocusing on he media ing oles o o ganiza ional cul u e and leade ship, employee
au onomy and skill de elopmen , echnos ess managemen , and e hical go e nance. The empi ical analysis con i med ha AI
adop ion has bo h di ec and indi ec e ec s on p oduc i i y. While he di ec impac (β = 0.41) demons a es ha AI in eg a ion
op imizes wo k lows and imp o es ask e iciency, he indi ec e ec s (β = 0.59) highligh he indispensable ole o human-cen ic
media o s. A suppo i e cul u e, pa icipa i e leade ship, and anspa en go e nance p ac ices signi ican ly ampli y AI’s
p oduc i i y bene i s. Con e sely, unmanaged au oma ion and echnos ess pa ially weaken hese posi i e e ec s. The s udy
ea i ms ha AI adop ion alone does no gua an ee pe o mance imp o emen . P oduc i i y gains ma e ialize only when AI is
embedded wi hin an o ganiza ional en i onmen ha os e s lea ning, e hical accoun abili y, and empowe men . These indings
con ibu e o socio- echnical sys ems and human–AI collabo a ion heo y by es ablishing ha he op imal ou comes eme ge om
balanced in e ac ion be ween echnology and human agency [6]–[8].
6.2 Manage ial Recommenda ions
Based on he indings, he ollowing p ac ical ecommenda ions a e p oposed o SME manage s, leade s, and policymake s
seeking o maximize he p oduc i i y bene i s o AI in eg a ion:
1. De elop Human-Cen ic AI S a egies: AI implemen a ion should be designed o augmen human decision-making, no
eplace i . Manage s should clea ly communica e how AI suppo s employees’ goals o mi iga e anxie y and esis ance
[7], [8].
2. P omo e T ans o ma ional Leade ship and Lea ning Cul u e: Encou aging expe imen a ion and con inuous imp o emen
os e s employee con idence in AI sys ems. Leade ship should ewa d inno a ion and p omo e collec i e lea ning h ough
c oss- unc ional collabo a ion [1], [10].
3. Enhance Digi al Li e acy and Reskilling: Con inuous AI li e acy p og ams mus be ins i u ionalized o imp o e wo k o ce
adap abili y. Employees wi h digi al skills a e be e able o use AI ools e ec i ely [3], [9].
4. Implemen E hical and T anspa en Go e nance: E hical AI deploymen , da a p i acy assu ance, and explainable AI
models build us and engagemen . Go e nance policies should de ine da a-use bounda ies and ensu e ai ness in
au oma ed decision-making p ocesses [11].
5. Moni o and Manage Technos ess: Regula employee eedback mechanisms should be es ablished o assess wo kload
balance and s ess le els. Flexible scheduling and human o e sigh in au oma ion can p e en bu nou [5], [6].
6. Fos e Feedback-Based Rein o cemen : P oduc i i y me ics de i ed om AI sys ems should in o m u u e adop ion
cycles. This eedback loop encou ages i e a i e lea ning and ensu es con inuous pe o mance imp o emen [8], [9], [10].
In e na ional Jou nal o Eme ging Resea ch in Science, Enginee ing, and Managemen
Vol. 1, Issue 4, pp.09-18, Oc obe 2025.
www.ije sem.com eISSN – 3107-9075
IJERSEM@2025 h ps://doi.o g/10.58482/ije sem. 1i4.2 17
6.3 Policy Implica ions
F om a policy pe spec i e, he esul s highligh he need o inclusi e digi al ans o ma ion amewo ks ha ex end AI access,
li e acy, and in as uc u e o SMEs. Go e nmen s and indus y associa ions should:
• P o ide incen i es o e hical and esponsible AI deploymen .
• Suppo sec o -speci ic AI adop ion oolki s and open da a pla o ms.
• Facili a e public–p i a e pa ne ships o SME wo k o ce aining.
Such ini ia i es would democ a ize AI adop ion and p e en p oduc i i y inequali y be ween digi ally ma u e and nascen
en e p ises.
6.4 Limi a ions and Fu u e Resea ch Di ec ions
Al hough he s udy o e s aluable insigh s, se e al limi a ions open a enues o u u e esea ch:
1. C oss-sec ional Design: The da a we e collec ed a a single poin in ime; u u e longi udinal s udies could cap u e
e ol ing p oduc i i y pa e ns o e ex ended AI adop ion phases.
2. Sel -Repo ed P oduc i i y: Fu u e wo k should in eg a e objec i e p oduc i i y me ics (e.g., ask comple ion a es,
pe o mance analy ics) alongside pe cep ual da a o imp o e accu acy.
3. Sec o al and Regional Va ia ions: Compa a i e s udies ac oss indus ies o coun ies can explo e how ins i u ional suppo
and echnological ma u i y shape AI-p oduc i i y dynamics.
4. In eg a ion o Hyb id Human–AI Wo k Models: Fu he esea ch should examine op imal ask di ision be ween humans
and AI, de eloping amewo ks o sha ed cogni ion and adap i e collabo a ion [7], [8].
5. Inclusion o Eme ging AI Technologies: Fu u e analyses may include gene a i e AI, adap i e lea ning models, and
mul imodal sys ems o assess how hese nex -gene a ion ools impac c ea i i y and knowledge wo k.
6.5 Final Rema ks
This s udy es ablishes ha AI adop ion in SMEs enhances employee p oduc i i y only when implemen ed e hically,
anspa en ly, and wi h adequa e human empowe men . The p oposed model demons a es ha o ganiza ional cul u e, leade ship,
and employee capabili y de elopmen ac as he p ima y condui s h ough which AI ansla es in o measu able pe o mance
ou comes. Fo SMEs aspi ing o h i e in he digi al economy, he key lies no me ely in adop ing AI—bu in adop ing i
esponsibly, collabo a i ely, and humanely.
FUNDING INFORMATION
This esea ch ecei ed no speci ic g an om any unding agency in he public, comme cial, o no - o -p o i sec o s.
ETHICS STATEMENT
This s udy did no in ol e human o animal subjec s and, he e o e, did no equi e e hical app o al.
STATEMENT OF CONFLICT OF INTERESTS
The au ho s decla e ha hey ha e no con lic s o in e es ela ed o his s udy.
LICENSING
This wo k is licensed unde a C ea i e Commons A ibu ion 4.0 In e na ional License.
REFERENCES
[1] G. Kassa and Z. Wo ku, “Impac o a i icial in elligence on employee pe o mance and i m p oduc i i y,” Jou nal
o Business and Managemen Resea ch, ol. 9, no. 2, 2024, doi: 10.1016/j.jbm .2024.105431.
[2] S. Gup a, M. Kuma , and A. Raj, “AI eadiness and o ganiza ional agili y in SMEs: Media ing ole o digi al
ans o ma ion,” In e na ional Jou nal o In o ma ion Managemen , ol. 77, no. 1, 2024, doi:
10.1016/j.ijin omg .2024.103788.
[3] X. Chen, Y. Li, and W. Zhou, “Human–AI collabo a ion and employee p oduc i i y: E idence om manu ac u ing
SMEs,” Technological Fo ecas ing and Social Change, ol. 201, 2024, doi: 10.1016/j. ech o e.2024.123567.
[4] J. P. Da is and M. C. Campos, “AI-enabled wo k low op imiza ion and c ea i e pe o mance,” In o ma ion Sys ems
F on ie s, ol. 27, no. 3, 2025, doi: 10.1007/s10796-025-10431-y.
[5] L. Rich e and M. Unge , “Technos ess and human capi al challenges in AI-in eg a ed wo kplaces,” Compu e s in
Human Beha io , ol. 151, 2024, doi: 10.1016/j.chb.2024.107210.
