Copy igh © ISRG Publishe s. All igh s Rese ed.
DOI: 10.5281/zenodo.17330694
267
ISRG PUBLISHERS
Abb e ia ed Key Ti le: ISRG J A s Humani Soc Sci
ISSN: 2583-7672 (Online)
Jou nal homepage: h ps://is gpublishe s.com/is gjahss
Volume – III Issue -V (Sep embe -Oc obe ) 2025
F equency: Bimon hly
The Impac o Da a P i acy Awa eness on AI-Powe ed Pe sonalized Ma ke ing and
Consume Beha io
Mahshid Asadollahi1*, Mohammad Akba i Asl2
1 Facul y o Business Adminis a ion, Depa men o Managemen And Law, To e ga a Uni e si y, Rome, I aly
2 Facul y o ou ism s a egy, cul u al he i age and made in i aly, Depa men o His o y, Humani ies and Socie y, To
e ga a Uni e si y, Rome, I aly
| Recei ed: 05.10.2025 | Accep ed: 09.10.2025 | Published: 12.10.2025
*Co esponding au ho : Mahshid Asadollahi
Facul y o Business Adminis a ion, Depa men o Managemen And Law, To e ga a Uni e si y, Rome, I aly
Abs ac
This s udy in es iga es he impac o AI-powe ed pe sonaliza ion on consume beha io , ocusing on engagemen , sa is ac ion, and
pu chase in en ion, while conside ing he media ing ole o us and he mode a ing in luence o da a p i acy awa eness. D awing
on a posi i is , deduc i e app oach, a c oss-sec ional su ey was conduc ed among 388 consume s wi h expe ience using AI-d i en
pe sonaliza ion on digi al pla o ms. The da a we e analyzed using S uc u al Equa ion Modeling (SEM) wi h Sma PLS 4.0. The
esul s e eal ha AI-powe ed pe sonaliza ion signi ican ly enhances consume engagemen , sa is ac ion, and pu chase in en ion.
T us , engagemen , and sa is ac ion ac as media ing mechanisms ha s eng hen he ela ionship be ween pe sonaliza ion and
pu chase ou comes. Howe e , da a p i acy awa eness was ound o mode a e hese e ec s: consume s wi h highe p i acy
awa eness demons a ed weake sa is ac ion and pu chase in en ions in esponse o pe sonaliza ion, whe eas hose less conce ned
abou p i acy esponded mo e posi i ely. In e es ingly, p i acy awa eness did no signi ican ly mode a e he ela ionship be ween
pe sonaliza ion and engagemen , sugges ing ha engagemen may s ill occu e en when p i acy conce ns a e ele a ed. The s udy
con ibu es heo e ically by in eg a ing da a p i acy awa eness as a bounda y condi ion in pe sonaliza ion esea ch and
empi ically alida ing he ole o us , engagemen , and sa is ac ion as key media o s. P ac ically, he indings highligh he need
o o ganiza ions o design pe sonaliza ion s a egies ha a e anspa en , us wo hy, and adap able o a ying le els o
consume p i acy awa eness. Policy implica ions emphasize he impo ance o clea egula o y amewo ks and esponsible AI
s anda ds o balance pe sonaliza ion bene i s wi h consume da a igh s.
Keywo ds: AI-powe ed pe sonaliza ion, consume us , engagemen , sa is ac ion, pu chase in en ion, da a p i acy awa eness
Copy igh © ISRG Publishe s. All igh s Rese ed.
DOI: 10.5281/zenodo.17330694
268
1. In oduc ion
A i icial In elligence (AI) has mo ed om he pe iphe y o
expe imen al echnologies in o he e y hea o con empo a y
ma ke ing p ac ice [1]. I s abili y o p ocess as amoun s o
consume da a in eal ime allows i ms o design highly a ge ed,
con ex -speci ic, and adap i e pe sonaliza ion s a egies. AI-
powe ed pe sonaliza ion e e s o he applica ion o algo i hms,
p edic i e analy ics, and machine lea ning echniques o deli e
indi idualized ecommenda ions, dynamic ad e isemen s, and
in e ac i e se ices ailo ed o he unique p e e ences o each
consume [2]. This app oach su passes adi ional segmen a ion
s a egies by de ec ing nuanced beha io al pa e ns and adap ing o
e ol ing cus ome needs, enabling companies o enhance
engagemen , educe esou ce ine iciencies, and os e s onge
b and–consume ela ionships [3]. Fo ins ance, e-comme ce
pla o ms such as Amazon and Alibaba le e age AI sys ems o
gene a e p oduc sugges ions, while s eaming se ices like Ne lix
and Spo i y cu a e indi idualized en e ainmen expe iences. The
dynamic lea ning capaci y o hese sys ems c ea es he imp ession
ha b ands ―unde s and‖ hei cus ome s, es ablishing
pe sonaliza ion as a s a egic o ganiza ional capabili y and a sou ce
o compe i i e ad an age [4,5].
The p oli e a ion o AI pe sonaliza ion has also eshaped consume
beha io in he digi al e a. Today’s consume s a e no passi e
ecipien s o ma ke ing con en bu ac i e pa icipan s in
algo i hmically media ed in e ac ions ha in luence hei
pe cep ions, decision-making, and loyal y. Pe sonalized
ecommenda ions alle ia e in o ma ion o e load, c ea e
con enience, and enhance ele ance, esul ing in highe
sa is ac ion and s onge pu chase in en ions [6]. F om a
psychological pe spec i e, pe sonaliza ion os e s us and
emo ional connec ion by making consume s eel ecognized and
alued. Resea ch shows ha such pe cei ed ele ance and
indi idual a en ion con ibu e o long- e m loyal y and mo e
posi i e b and e alua ions. A he same ime, howe e ,
pe sonaliza ion can p o oke ambi alen eac ions. O e - a ge ing
and in usi e messaging o en aise conce ns abou manipula ion o
su eillance, leading o skep icism and esis ance. This duali y—
whe e pe sonaliza ion simul aneously deligh s and unse les—
unde sco es he need o examine no only he ou comes o
pe sonaliza ion bu also he unde lying condi ions ha shape
consume esponses [7].
Cen al o hese condi ions is he g owing ele ance o da a p i acy
awa eness. AI-powe ed pe sonaliza ion depends on he con inuous
collec ion and p ocessing o pe sonal in o ma ion, anging om
ansac ional his o ies o eal- ime geoloca ion da a. While hese
capabili ies c ea e unp eceden ed oppo uni ies o ele ance and
con enience, hey also aise p essing conce ns abou anspa ency,
ai ness, and secu i y. Consume s a e inc easingly conscious o
how hei da a is collec ed and used, and high-p o ile cases o
misuse ha e ampli ied hei sensi i i y o p i acy isks. P i acy
awa eness now ope a es as a cen al de e minan o consume us ,
shaping whe he pe sonaliza ion is in e p e ed as empowe ing o
in asi e [8,9]. Regula o y amewo ks such as he Gene al Da a
P o ec ion Regula ion (GDPR) and he Cali o nia Consume
P i acy Ac (CCPA) ha e u he heigh ened expec a ions by
ensh ining consume igh s o e consen , access, and da a con ol.
In his con ex , companies a e no longe e alua ed solely on he
quali y o hei pe sonalized o e ings bu also on he in eg i y and
esponsibili y wi h which hey handle consume da a [10].
Despi e he clea impo ance o his issue, a signi ican esea ch
gap emains. Exis ing s udies ha e p ima ily ocused on he
echnical e iciency o pe sonaliza ion and i s di ec e ec s on
sa is ac ion, engagemen , o pu chase in en ions [4,7-9].
Compa a i ely less a en ion has been de o ed o he mode a ing
ole o p i acy awa eness in hese ela ionships. C i ical ques ions
emain unanswe ed: Do consume s wi h high p i acy awa eness
equi e s onge assu ances o secu i y and anspa ency be o e
hey accep pe sonaliza ion? Con e sely, does low p i acy
awa eness ampli y posi i e pe cep ions by minimizing pe cei ed
isks? Fu he mo e, while us is widely ecognized as a media o
o pe sonaliza ion ou comes, he ex en o which p i acy awa eness
in luences us i sel is unde explo ed. Add essing hese ques ions
is essen ial o ad ancing bo h heo y and p ac ice.
Acco dingly, he p esen s udy in es iga es he ole o da a p i acy
awa eness in shaping consume esponses o AI-powe ed
pe sonaliza ion. Speci ically, i examines (i) he di ec impac o
pe sonaliza ion on consume engagemen , sa is ac ion, and
pu chase in en ions; (ii) he media ing ole o us in he
pe sonaliza ion–beha io ela ionship; and (iii) he mode a ing
in luence o p i acy awa eness on he s eng h and di ec ion o
hese e ec s. By in eg a ing da a p i acy awa eness in o he
pe sonaliza ion–beha io amewo k, his s udy con ibu es o he
li e a u e on AI ma ke ing and consume beha io while o e ing
ac ionable insigh s o i ms seeking o balance pe sonaliza ion
e ec i eness wi h esponsible da a p ac ices. Resea ch Ques ions
a e as ollows:
1. How does AI-powe ed pe sonaliza ion in luence
consume beha io ou comes such as engagemen ,
sa is ac ion, and pu chase in en ions?
2. Wha ole does consume us play in media ing he
ela ionship be ween AI-powe ed pe sonaliza ion and
beha io al ou comes?
3. How does da a p i acy awa eness a ec consume
pe cep ions o AI-powe ed pe sonaliza ion?
4. To wha ex en does da a p i acy awa eness mode a e he
ela ionship be ween AI-powe ed pe sonaliza ion and
consume beha io ou comes?
5. A e he e signi ican di e ences in consume esponses
o pe sonaliza ion be ween indi iduals wi h high e sus
low le els o da a p i acy awa eness?
2. Li e a u e Re iew
1. A i icial In elligence as a C oss-Indus y Enable
A i icial in elligence (AI) has demons a ed ans o ma i e
po en ial ac oss di e se domains, p o iding a ounda ion o
unde s anding i s ole in ma ke ing and consume con ex s. In
ene gy and anspo a ion, AI-d i en op imiza ion echniques ha e
been employed o elec ic ehicle cha ging managemen , pa king
lo alloca ion, and sma g id ene gy balancing, showing how
in elligen algo i hms can enhance eliabili y and decision-making
in dynamic en i onmen s [11-14]. In heal hca e, ad anced deep
lea ning models such as imp o ed CNNs and U-Ne a chi ec u es
ha e been applied o au ism diagnosis and umo de ec ion, while
usion-based explainable AI app oaches suppo anspa ency in
medical decision-making [15-17]. AI applica ions ha e also
ex ended o u ban sys ems, including uzzy logic models o a ic
densi y p edic ion [18], se ice-o ien ed a chi ec u es o secu e
da a hiding [19], and geospa ial analy ics o unde s anding c ime
Copy igh © ISRG Publishe s. All igh s Rese ed.
DOI: 10.5281/zenodo.17330694
269
pa e ns in u ban ne wo ks [20]. In he ield o enginee ing and
sa e y, AI has been in eg a ed in o explainable and in e p e able
machine lea ning o geological haza d p edic ion, in elligen
o ecas ing, and ea ly wa ning sys ems [21-24]. Fu he mo e,
hyb id in o ma ion usion app oaches, physics-in o med aul
diagnos ics, and ans e able ea u e lea ning ha e been de eloped
o imp o e he moni o ing and eliabili y o complex indus ial
sys ems [25-27]. Collec i ely, hese ad ances illus a e AI’s
e sa ili y in sol ing p oblems ha in ol e unce ain y, scale, and
in e p e abili y, es ablishing a echnological backd op o i s
adop ion in ma ke ing h ough pe sonaliza ion and consume
engagemen .
2. AI-Powe ed Pe sonalized Ma ke ing: Concep s and
Applica ions
AI-powe ed pe sonaliza ion has apidly eme ged as one o he
mos ans o ma i e applica ions o a i icial in elligence in
business p ac ice. Unlike adi ional one-size- i s-all app oaches,
pe sonaliza ion uses algo i hms, p edic i e analy ics, and machine
lea ning models o analyze consume da a and ailo in e ac ions o
indi idual p e e ences in eal ime. Recommende sys ems,
dynamic p icing ools, na u al language–enabled cha bo s, and
au oma ed cus ome se ice agen s exempli y how pe sonaliza ion
pe mea es di e se indus ies anging om e ail and a el o
inance and en e ainmen [7,8].
Schola s a gue ha pe sonaliza ion achie es a dual unc ion: i
enhances consume ele ance while simul aneously c ea ing
ope a ional e iciency o i ms. By an icipa ing consume needs
and s eamlining decision-making, companies educe was ed
ma ke ing expendi u e while inc easing consume engagemen .
Indus y leade s such as Amazon, Ne lix, and Spo i y ha e
success ully ope a ionalized pe sonaliza ion engines o deli e
cu a ed expe iences ha ein o ce b and loyal y. AI’s abili y o
con inuously lea n and adap u he deepens pe sonaliza ion, as
sys ems e ol e in line wi h shi ing consume p e e ences [28].
A he same ime, pe sonaliza ion equi es signi ican in es men s
in da a in as uc u e, ad anced analy ics, and o ganiza ional
in eg a ion. S udies highligh ha i ms wi h highe le els o digi al
ma u i y achie e g ea e e u ns om pe sonaliza ion ini ia i es
compa ed o i ms wi h agmen ed digi al capabili ies [29].
The e o e, AI-powe ed pe sonaliza ion is no simply a ma ke ing
echnique bu a dynamic capabili y ha ede ines compe i i e
ad an age in digi al economies.
3. Consume Beha io and Pe sonaliza ion Ou comes
The e ec s o pe sonaliza ion on consume beha io a e well
documen ed in ma ke ing li e a u e. Pe sonalized expe iences
educe cogni i e e o by il e ing i ele an in o ma ion, making
he decision p ocess mo e e icien and enjoyable. Resea ch
consis en ly demons a es ha pe sonaliza ion enhances
sa is ac ion, pu chase in en ion, and loyal y. Consume s o en
pe cei e pe sonaliza ion as e idence ha b ands unde s and and
alue hem, he eby c ea ing s onge emo ional connec ions ha
ex end beyond ansac ional in e ac ions [30].
Ye , he in luence o pe sonaliza ion is no uni o mly posi i e.
Se e al s udies emphasize he ambi alence in consume esponses.
On one hand, pe sonaliza ion can gene a e deligh by su p ising
consume s wi h ele an , imely, and use ul ecommenda ions. On
he o he hand, o e -pe sonaliza ion o hype - a ge ed messaging
can elici discom o and skep icism, pa icula ly when consume s
pe cei e ha i ms know ― oo much‖ abou hei p i a e li es. This
phenomenon, o en desc ibed as he ―c eepiness ac o ,‖ highligh s
he delica e balance be ween alue c ea ion and pe cei ed in usion
[31].
Fu he mo e, consume eac ions a e in luenced by psychological,
cul u al, and con ex ual a iables. Some consume s emb ace
pe sonaliza ion due o i s con enience, while o he s esis i due o
pe cei ed manipula ion o au onomy loss. Consequen ly,
pe sonaliza ion ou comes canno be ully unde s ood wi hou
conside ing b oade consume a i udes owa d da a use and
echnological media ion. This unde sco es he impo ance o
in eg a ing cons uc s such as us , p i acy awa eness, and digi al
con idence in o pe sonaliza ion esea ch [7].
4. Da a P i acy Awa eness and Consume T us
In pa allel wi h he ise o pe sonaliza ion, consume awa eness o
da a p i acy has in ensi ied. High-p o ile da a b eaches, g owing
conce ns o e su eillance, and inc eased media co e age o
une hical da a p ac ices ha e heigh ened consume sensi i i y o
how pe sonal in o ma ion is collec ed, s o ed, and sha ed. P i acy
awa eness encompasses consume s’ knowledge o da a p ac ices,
hei pe cep ions o associa ed isks, and hei expec a ions o
o ganiza ional accoun abili y [8].
T us plays a pi o al ole in his dynamic. When consume s belie e
ha i ms handle hei da a esponsibly, hey a e mo e likely o
engage wi h pe sonalized se ices, sha e addi ional in o ma ion,
and o m long- e m ela ionships. Con e sely, a lack o
anspa ency o secu i y unde mines us , leading o a oidance,
nega i e wo d o mou h, o wi hd awal om digi al engagemen .
Regula o y amewo ks such as he Gene al Da a P o ec ion
Regula ion (GDPR) and he Cali o nia Consume P i acy Ac
(CCPA) ha e ein o ced his dynamic by g an ing consume s
explici igh s, including in o med consen , da a po abili y, and he
igh o e asu e. These policies ha e bo h empowe ed consume s
and inc eased hei expec a ions o i ms’ e hical beha io [10].
Resea ch shows ha p i acy awa eness can bo h acili a e and
hinde pe sonaliza ion adop ion. Fo some consume s, awa eness
o s ong p i acy sa egua ds enhances con idence and us ,
encou aging engagemen wi h AI-d i en pe sonaliza ion. Fo
o he s, heigh ened awa eness ampli ies skep icism and leads o
isk-a e se beha io . Unde s anding his dual ole o p i acy
awa eness is c ucial o i ms seeking o balance pe sonaliza ion
bene i s wi h consume com o [8,9].
5. The Mode a ing Role o P i acy Awa eness in AI-
Pe sonaliza ion
While pe sonaliza ion and us ha e been ex ensi ely s udied,
ewe in es iga ions ha e posi ioned p i acy awa eness as a
bounda y condi ion ha mode a es he ela ionship be ween
pe sonaliza ion and consume ou comes. A g owing body o
li e a u e sugges s ha he e ec s o pe sonaliza ion depend
hea ily on consume p i acy o ien a ion. Fo consume s wi h high
p i acy awa eness, pe sonaliza ion is e alua ed c i ically: hey
demand clea e idence o da a p o ec ion, anspa en
communica ion, and meaning ul consen be o e engaging
posi i ely wi h pe sonalized o e ings. In con as , consume s wi h
low p i acy awa eness o lowe sensi i i y o da a isks a e mo e
likely o pe cei e pe sonaliza ion bene i s di ec ly, wi hou
heigh ened conce ns abou su eillance o manipula ion [32].
This mode a ing ole has signi ican implica ions o bo h heo y
and p ac ice. Theo e ically, i sugges s ha pe sonaliza ion
ou comes canno be gene alized ac oss popula ions; a he , hey
Copy igh © ISRG Publishe s. All igh s Rese ed.
DOI: 10.5281/zenodo.17330694
270
a y based on he deg ee o consume p i acy awa eness.
P ac ically, i highligh s he impo ance o con ex -sensi i e
s a egies: i ms may need o adap hei pe sonaliza ion ac ics
depending on whe he hei a ge audience is cha ac e ized by high
o low p i acy sensi i i y. Fo example, in ma ke s wi h s onge
egula o y en i onmen s and heigh ened consume awa eness,
anspa ency and e hical posi ioning may be as impo an as
pe sonaliza ion e ec i eness. Con e sely, in ma ke s wi h lowe
awa eness, he emphasis may emain on deli e ing con enience
and ele ance. By concep ualizing p i acy awa eness as a
mode a o , his s udy add esses a c i ical gap in pe sonaliza ion
esea ch. I highligh s he necessi y o in eg a ing consume -le el
psychological and e hical conside a ions in o models o AI-
powe ed ma ke ing e ec i eness.
6. Concep ual F amewo k and Hypo heses De elopmen
Syn hesizing insigh s om he e iewed li e a u e, he s udy
p oposes a concep ual amewo k ha posi ions AI-powe ed
pe sonaliza ion as a d i e o consume beha io ou comes,
including engagemen , sa is ac ion, and pu chase in en ions (Figu e
1). Consume us is iden i ied as a key media o , e lec ing he
mechanism h ough which pe sonaliza ion ansla es in o posi i e
beha io al esponses. C ucially, da a p i acy awa eness is
in oduced as a mode a o , condi ioning he s eng h and di ec ion
o he pe sonaliza ion–ou come ela ionship.
This amewo k builds on he in e sec ion o pe sonaliza ion, us ,
and p i acy esea ch. I assumes ha while AI pe sonaliza ion has
he po en ial o enhance consume expe iences, i s e ec i eness is
con ingen upon consume s’ le el o p i acy awa eness. By
examining hese in e ac ions, he s udy ex ends exis ing heo y,
esponding o calls o mo e nuanced unde s anding o
pe sonaliza ion ou comes in da a-d i en en i onmen s [33].
Figu e 1. Theo e ical F amewo k
7. Signi icance o he S udy
The signi icance o his s udy lies in i s ocus on he in e sec ion
be ween echnological inno a ion and consume e hics in digi al
ma ke ing. AI-powe ed pe sonaliza ion has become a
ans o ma i e o ce in shaping cus ome expe iences, imp o ing
sa is ac ion, and enhancing pu chase in en ions. Ye , i s success
c i ically depends on consume s’ us and hei awa eness o how
pe sonal da a is collec ed and used. By in es iga ing he ole o
da a p i acy awa eness, his s udy add esses one o he mos
p essing challenges in he digi al economy: balancing
pe sonaliza ion bene i s wi h e hical and anspa en da a p ac ices.
This esea ch is impo an o businesses seeking o enhance
pe sonaliza ion s a egies wi hou comp omising consume us . I
also p o ides meaning ul insigh s o policymake s conce ned wi h
digi al igh s, consume p o ec ion, and egula o y amewo ks, as
well as o he gene al public na iga ing pe sonalized online
en i onmen s. In doing so, he s udy con ibu es o a mo e
comp ehensi e unde s anding o he oppo uni ies and isks ha
accompany AI-d i en pe sonaliza ion in con empo a y ma ke ing.
8. Theo e ical Con ibu ion
Ex ends pe sonaliza ion esea ch by in oducing da a
p i acy awa eness as a mode a ing a iable, he eby
adding a c i ical bounda y condi ion o es ablished
models o pe sonaliza ion and consume beha io .
Enhances heo e ical unde s anding o he media ing ole
o consume us in ansla ing pe sonaliza ion e o s
in o beha io al ou comes.
B idges pe spec i es om ma ke ing, consume
psychology, and digi al e hics, c ea ing an in eg a ed
Copy igh © ISRG Publishe s. All igh s Rese ed.
DOI: 10.5281/zenodo.17330694
271
amewo k ha explains bo h posi i e and nega i e
esponses o AI-powe ed pe sonaliza ion.
9. P ac ical Con ibu ion
P o ides ac ionable insigh s o ma ke e s on how o
design AI-d i en pe sonaliza ion s a egies ha enhance
consume expe ience while espec ing p i acy
expec a ions.
O e s guidance o i ms on building digi al us by
in eg a ing anspa ency, accoun abili y, and e hical da a
managemen in o pe sonaliza ion p ac ices.
Sugges s con ex -sensi i e s a egies ha can be adap ed
o audiences wi h a ying le els o p i acy awa eness,
enabling i ms o maximize pe sonaliza ion e ec i eness
in di e en ma ke s.
10. Me hodological Con ibu ion
P oposes an inno a i e SEM-based amewo k o
analyze he complex ela ionships among
pe sonaliza ion, us , p i acy awa eness, and consume
beha io ou comes.
Employs a combina ion o su eys and in e iews,
p o iding bo h quan i a i e igo and quali a i e dep h in
cap u ing consume pe cep ions.
Inco po a es mode a ion and media ion analysis o
disen angle he condi ional and indi ec e ec s o
pe sonaliza ion, o e ing me hodological cla i y o
u u e s udies in digi al ma ke ing.
11. Jus i ica ion
This esea ch is imely and essen ial, gi en he accele a ing
in eg a ion o AI in o ma ke ing ope a ions and he pa allel ise in
consume p i acy awa eness. While exis ing s udies demons a e
ha pe sonaliza ion enhances pu chase in en ions h ough
consume engagemen , sa is ac ion, and pe cei ed alue, a ewe
ha e conside ed how p i acy awa eness condi ions hese e ec s.
The g owing socie al emphasis on digi al igh s, anspa ency, and
e hical ma ke ing unde sco es he impo ance o s udying p i acy
as a de e minan o consume accep ance.
By examining he in e play be ween pe sonaliza ion, us , and
p i acy awa eness, his s udy no only ills a c i ical gap in he
li e a u e bu also gene a es indings wi h di ec implica ions o
ma ke ing p ac ice and policy. The ou comes o his esea ch will
bene i schola s seeking o ad ance heo y, businesses s i ing o
gain consume us , and policymake s asked wi h ensu ing
esponsible use o AI in consume ma ke s.
12. Hypo heses
Di ec E ec s
H1: AI-powe ed pe sonaliza ion has a posi i e and
signi ican e ec on consume engagemen .
H2: AI-powe ed pe sonaliza ion has a posi i e and
signi ican e ec on consume sa is ac ion.
H3: AI-powe ed pe sonaliza ion has a posi i e and
signi ican e ec on consume pu chase in en ions.
H4: Consume us has a posi i e and signi ican e ec
on consume pu chase in en ions.
H5: Consume engagemen has a posi i e and signi ican
e ec on consume sa is ac ion.
H6: Consume sa is ac ion has a posi i e and signi ican
e ec on consume pu chase in en ions.
Media ing E ec s
H7: Consume us media es he ela ionship be ween
AI-powe ed pe sonaliza ion and pu chase in en ions.
H8: Consume us media es he ela ionship be ween
AI-powe ed pe sonaliza ion and consume sa is ac ion.
H9: Consume engagemen media es he ela ionship
be ween AI-powe ed pe sonaliza ion and pu chase
in en ions.
H10: Consume sa is ac ion media es he ela ionship
be ween AI-powe ed pe sonaliza ion and pu chase
in en ions.
Mode a ing E ec s
H11: Da a p i acy awa eness mode a es he ela ionship
be ween AI-powe ed pe sonaliza ion and consume
engagemen , such ha he ela ionship is weake when
p i acy awa eness is high.
H12: Da a p i acy awa eness mode a es he ela ionship
be ween AI-powe ed pe sonaliza ion and consume
sa is ac ion, such ha he ela ionship is weake when
p i acy awa eness is high.
H13: Da a p i acy awa eness mode a es he ela ionship
be ween AI-powe ed pe sonaliza ion and consume
pu chase in en ions, such ha he ela ionship is weake
when p i acy awa eness is high.
H14: Da a p i acy awa eness mode a es he indi ec
ela ionship be ween AI-powe ed pe sonaliza ion and
pu chase in en ions h ough consume us .
3. Me hodology
3.1 Resea ch Philosophy and App oach
This s udy adop s a posi i is esea ch philosophy and ollows a
deduc i e app oach, consis en wi h [34], who emphasizes ha a
deduc i e design is mos app op ia e when he objec i e is o es a
se o p ede ined hypo heses. By applying his app oach, he
esea ch seeks o empi ically con i m o ejec hypo hesized
ela ionships conce ning he impac o AI-powe ed pe sonaliza ion,
consume us , and da a p i acy awa eness on consume beha io
ou comes.
3.2 Resea ch Design
The s udy employs a Mono Me hod Quan i a i e (MMQ) s a egy,
using s uc u ed su ey ques ionnai es as he sole da a collec ion
ins umen [35]. A c oss-sec ional design is applied, as he equi ed
da a a e collec ed om all pa icipan s a a single poin in ime.
This design is app op ia e o examining pe cep ions and beha io s
ela ed o pe sonaliza ion, p i acy awa eness, and pu chase
in en ions in he con ex o digi al ma ke ing.
3.3 Popula ion and Sampling
The a ge popula ion o his s udy consis s o consume s who
egula ly use digi al pla o ms o shopping and online pu chasing
ac i i ies. To ensu e ha pa icipan s ha e adequa e exposu e o
AI-d i en pe sonaliza ion, pu posi e sampling is adop ed. This
Copy igh © ISRG Publishe s. All igh s Rese ed.
DOI: 10.5281/zenodo.17330694
272
echnique allows he selec ion o esponden s who a e amilia wi h
pe sonalized ecommenda ions, a ge ed ad e ising, o AI-enabled
cus ome in e aces [35].
Based on s a is ical guidelines o S uc u al Equa ion Modeling
(SEM), he ecommended sample size should be a leas en imes
he numbe o s uc u al pa hs leading o a cons uc . Conside ing
he p oposed model includes mul iple di ec , media ing, and
mode a ing ela ionships, a sample size o 388 consume s is
deemed su icien o obus analysis.
3.4 Da a Collec ion Ins umen
A s uc u ed ques ionnai e will be designed wi h i e-poin Like
scale i ems ( anging om 1 = S ongly Disag ee o 5 = S ongly
Ag ee). The ques ionnai e is di ided in o sec ions measu ing:
1. AI-Powe ed Pe sonaliza ion (e.g., pe cei ed ele ance,
equency o ecommenda ions, pe cei ed use ulness).
2. Consume T us (e.g., pe cei ed secu i y, anspa ency,
eliabili y).
3. Consume Engagemen (e.g., a en ion, in e ac ion,
b and in ol emen ).
4. Consume Sa is ac ion (e.g., o e all sa is ac ion wi h
pe sonalized expe iences).
5. Pu chase In en ions (e.g., likelihood o pu chasing
based on pe sonalized ecommenda ions).
6. Da a P i acy Awa eness (e.g., awa eness o da a
collec ion p ac ices, pe cei ed isk, conce n o e da a
use).
The i ems will be adap ed om alida ed scales in p io esea ch
on pe sonaliza ion, us , and p i acy [9,10,32].
3.5 Da a Analysis Technique
The collec ed da a will be analyzed using S uc u al Equa ion
Modeling (SEM) wi h Sma PLS 4.0, a me hod chosen o i s
sui abili y in handling complex models ha inco po a e bo h
media ion and mode a ion e ec s [36]. The analysis will be ca ied
ou in wo main s ages. In he i s s age, he measu emen model
will be e alua ed o ensu e eliabili y and alidi y. Reliabili y will
be examined h ough C onbach’s Alpha and Composi e Reliabili y
(CR), while con e gen alidi y will be assessed using he A e age
Va iance Ex ac ed (AVE). Disc iminan alidi y will hen be
e alua ed using bo h he Fo nell–La cke c i e ion and he
He e o ai –Mono ai (HTMT) a io, he eby ensu ing ha each
cons uc is bo h in e nally consis en and empi ically dis inc .
In he second s age, he s uc u al model will be e alua ed o es
he hypo hesized ela ionships. Pa h coe icien s (β) and
signi icance le els (p- alues) will be ob ained h ough a
boo s apping p ocedu e wi h 5,000 esamples, p o iding obus
es ima es o s a is ical signi icance. The di ec e ec s o AI-
powe ed pe sonaliza ion on engagemen , sa is ac ion, and pu chase
in en ion will be es ed, alongside he media ing ole o consume
us , engagemen , and sa is ac ion using he indi ec e ec s
unc ion. Fu he mo e, he mode a ing in luence o da a p i acy
awa eness will be examined by in oducing in e ac ion e ms and
assessing hei e ec s on consume beha io ou comes. To assess
he explana o y and p edic i e powe o he model, he coe icien
o de e mina ion (R²) and p edic i e ele ance (Q²) will also be
epo ed. This wo-s ep app oach ensu es ha bo h he
measu emen and s uc u al aspec s o he model a e igo ously
alida ed be o e d awing conclusions om he empi ical indings.
3.6 E hical Conside a ions
All esponden s will be in o med o he s udy’s academic pu pose,
and hei pa icipa ion will be olun a y. Anonymi y and
con iden iali y o esponses will be s ic ly main ained [36]. The
s udy will comply wi h da a p o ec ion egula ions such as GDPR
o ensu e e hical handling o pe sonal in o ma ion.
4. Resul s
The measu emen model was i s assessed o es ablish eliabili y
and alidi y o he cons uc s be o e es ing he hypo hesized
s uc u al ela ionships. Following Hai e al. (2022), h ee key
c i e ia we e examined: in e nal consis ency eliabili y, con e gen
alidi y, and disc iminan alidi y.
4.1 Reliabili y and Con e gen Validi y
Table 1 p esen s he esul s o eliabili y and con e gen alidi y
assessmen . C onbach’s Alpha alues anged om 0.873 (Pu chase
In en ion) o 0.915 (Consume T us ), exceeding he ecommended
h eshold o 0.70. Simila ly, Composi e Reliabili y (CR) alues
we e consis en ly abo e 0.87, suppo ing s ong cons uc
eliabili y. The A e age Va iance Ex ac ed (AVE) alues anged
om 0.591 o 0.730, well abo e he 0.50 benchma k. These esul s
demons a e ha he cons uc s cap u e su icien a iance om
hei indica o s and con i m con e gen alidi y.
Table 1: Reliabili y and Validi y
Cons uc
C onbach’s Alpha
Composi e Reliabili y (CR)
A e age Va iance Ex ac ed (AVE)
AI-Pe sonaliza ion
0.885
0.884
0.605
Consume T us
0.915
0.915
0.730
Consume Engagemen
0.879
0.878
0.591
Consume Sa is ac ion
0.905
0.905
0.704
Pu chase In en ion
0.873
0.873
0.632
Da a P i acy Awa eness
0.896
0.896
0.633
4.2 Disc iminan Validi y: Fo nell–La cke C i e ion
Disc iminan alidi y was assessed using he Fo nell–La cke
c i e ion. As shown in Table 2, he squa e oo o he AVE o each
cons uc (diagonal alues anging om 0.769 o 0.854) was
g ea e han i s co ela ions wi h o he cons uc s. Fo ins ance, he
squa e oo o AVE o Consume T us (0.854) exceeded i s
co ela ions wi h AI-Pe sonaliza ion (0.512) and Pu chase
In en ion (0.660). This indica es ha each cons uc sha es mo e
a iance wi h i s own indica o s han wi h o he cons uc s, hus
con i ming disc iminan alidi y.
Copy igh © ISRG Publishe s. All igh s Rese ed.
DOI: 10.5281/zenodo.17330694
273
Table 2: Disc iminan Validi y (Fo nell–La cke C i e ion)
Cons uc
AI-Pe sonaliza ion
Consume
T us
Consume
Engagemen
Consume
Sa is ac ion
Pu chase
In en ion
Da a P i acy
Awa eness
AI-Pe sonaliza ion
0.778
Consume T us
0.512
0.854
Consume Engagemen
0.685
0.447
0.769
Consume Sa is ac ion
0.620
0.417
0.612
0.839
Pu chase In en ion
0.580
0.660
0.406
0.650
0.795
Da a P i acy Awa eness
0.447
0.580
0.691
0.464
0.455
0.796
4.3 Disc iminan Validi y: HTMT C i e ion
To u he assess disc iminan alidi y, he He e o ai –Mono ai
Ra io (HTMT) was examined. Table 3 shows ha HTMT alues
anged om 0.547 o 0.781, all o which a e below he
conse a i e h eshold o 0.85 and well unde he mo e lenien
h eshold o 0.90. Fo example, he HTMT a io be ween
Consume Sa is ac ion and Pu chase In en ion was 0.781,
com o ably below he cu -o . These indings ein o ce he
conclusion ha he cons uc s a e empi ically dis inc and ee om
mul icollinea i y issues.
Table 3: HTMT (He e o ai –Mono ai Ra io)
Cons uc
AI-
Pe sonaliza ion
Consume T us
Consume
Engagemen
Consume
Sa is ac ion
Pu chase
In en ion
Da a P i acy
Awa eness
AI-Pe sonaliza ion
–
Consume T us
0.612
–
Consume Engagemen
0.648
0.701
–
Consume Sa is ac ion
0.672
0.724
0.759
–
Pu chase In en ion
0.583
0.692
0.733
0.781
–
Da a P i acy Awa eness
0.547
0.703
0.654
0.695
0.669
–
O e all, he esul s om C onbach’s Alpha, Composi e Reliabili y,
AVE, Fo nell–La cke c i e ion, and HTMT analysis p o ide
s ong e idence ha he measu emen model demons a es
accep able eliabili y, con e gen alidi y, and disc iminan
alidi y. Ha ing es ablished he obus ness o he measu emen
model, he s udy p oceeds o e alua e he s uc u al model and es
he p oposed hypo heses.
4.4 Indica o Reliabili y (Ou e Loadings)
Indica o eliabili y was assessed by examining he ou e loadings
o each measu emen i em on i s co esponding la en cons uc .
Ou e loadings ep esen he deg ee o co ela ion be ween an
obse ed indica o and he la en a iable i is in ended o measu e.
High loading alues indica e ha he indica o con ibu es s ongly
o he a iance o he cons uc , while low alues sugges weake
ep esen a ion. Acco ding o Hai e al. (2022), a loading o 0.70 o
abo e is gene ally ega ded as he benchma k, as his implies ha
a leas 50% o he a iance in he indica o is explained by he
la en cons uc . In his s udy, he majo i y o indica o s
demons a ed s ong loadings, exceeding he 0.70 h eshold and
he eby con i ming hei adequacy as eliable measu es. Fo
ins ance, he indica o s o Consume T us and Consume
Sa is ac ion consis en ly loaded be ween 0.804 and 0.879, which
e lec s a high le el o consis ency and measu emen accu acy.
Such s ong loadings sugges ha hese i ems p o ide a p ecise
ep esen a ion o he unde lying cons uc s and can be elied upon
in subsequen model es ing. A he same ime, a small numbe o
indica o s displayed ma ginally lowe loadings. Fo example, CE1
(0.697) and DPA1 (0.721) ell sligh ly below he ideal h eshold
bu emained wi hin he accep able ole ance ange (0.60–0.70). In
line wi h he ecommenda ions o Chin (1998) and Hai e al.
(2019), hese indica o s we e e ained o h ee easons: (i) hey
we e heo e ically signi ican in cap u ing aspec s o engagemen
and p i acy awa eness ha o he i ems did no ully co e , (ii) hei
inclusion did no educe he composi e eliabili y (CR) o A e age
Va iance Ex ac ed (AVE) below accep able le els, and (iii) hey
con ibu ed o he con en alidi y o hei espec i e cons uc s,
ensu ing ha he ull concep ual domain was ep esen ed.
The combina ion o s ong high-loading indica o s and ca e ully
jus i ied bo de line indica o s p o ides e idence o a obus
measu emen model. This demons a es ha he obse ed a iables
adequa ely cap u e hei espec i e la en cons uc s, ensu ing ha
he s udy’s indings a e bo h s a is ically eliable and heo e ically
g ounded.
Table 4: Ou e Loadings
Cons uc
Indica o
Ou e Loading
AI-Powe ed
Pe sonaliza ion
AP1
0.712
AP2
0.754
AP3
0.802
AP4
0.835
AP5
0.781
Consume T us
CT1
0.844
Copy igh © ISRG Publishe s. All igh s Rese ed.
DOI: 10.5281/zenodo.17330694
274
CT2
0.879
CT3
0.867
CT4
0.826
Consume Engagemen
CE1
0.697
CE2
0.742
CE3
0.803
CE4
0.816
CE5
0.781
Consume Sa is ac ion
CS1
0.826
CS2
0.853
CS3
0.871
CS4
0.804
Pu chase In en ion
PI1
0.759
PI2
0.782
PI3
0.801
PI4
0.836
Da a P i acy Awa eness
DPA1
0.721
DPA2
0.766
DPA3
0.832
DPA4
0.849
DPA5
0.803
4.5 Hypo heses Tes ing
The esul s o he hypo hesis es ing indica e ha AI-powe ed
pe sonaliza ion exe s signi ican posi i e e ec s on engagemen
(H1: T=5.379, p=0.000), sa is ac ion (H2: T=4.656, p=0.000), and
pu chase in en ion (H3: T=4.547, p=0.000). Simila ly, consume
us was ound o ha e a s ong and signi ican di ec e ec on
pu chase in en ion (H4: T=4.188, p=0.000), while engagemen
signi ican ly enhanced sa is ac ion (H5: T=5.661, p=0.000).
Sa is ac ion also displayed a posi i e and signi ican in luence on
pu chase in en ion (H6: T=4.343, p=0.000).
Media ion e ec s we e also alida ed. Consume us signi ican ly
media ed he ela ionship be ween pe sonaliza ion and pu chase
in en ion (H7: T=3.692, p=0.000) as well as be ween
pe sonaliza ion and sa is ac ion (H8: T=3.235, p=0.001).
Engagemen se ed as a signi ican media o be ween
pe sonaliza ion and pu chase in en ion (H9: T=3.764, p=0.000),
while sa is ac ion media ed he pe sonaliza ion–pu chase in en ion
pa h (H10: T=4.222, p=0.000).
The mode a ing ole o da a p i acy awa eness p oduced mixed
esul s. I s in e ac ion e ec weakened he ela ionship be ween
AI-pe sonaliza ion and sa is ac ion (H12: T=2.056, p=0.040) and
be ween pe sonaliza ion and pu chase in en ion (H13: T=2.282,
p=0.023). A simila mode a ing e ec was obse ed o he
indi ec ela ionship h ough us (H14: T=2.088, p=0.037).
Howe e , he mode a ion e ec on engagemen was no
s a is ically signi ican (H11: T=1.676, p=0.094).
O e all, hese esul s p o ide e idence ha AI-pe sonaliza ion
signi ican ly in luences consume beha io ou comes bo h di ec ly
and indi ec ly, while da a p i acy awa eness ac s as a bounda y
condi ion, weakening some o hese ela ionships when awa eness
le els a e high.
Table 5: Hypo heses Resul s
Hypo hesis
Pa h Rela ionship
O iginal
Sample
(O)
Sample
Mean
(M)
S anda d
De ia ion
(STD)
T S a is ics
(O/STD)
P Values
H1
AI-Pe sonaliza ion → Engagemen
0.312
0.315
0.058
5.379
0.000
H2
AI-Pe sonaliza ion → Sa is ac ion
0.284
0.288
0.061
4.656
0.000
H3
AI-Pe sonaliza ion → Pu chase In en ion
0.241
0.245
0.053
4.547
0.000
H4
Consume T us → Pu chase In en ion
0.268
0.271
0.064
4.188
0.000
H5
Engagemen → Sa is ac ion
0.334
0.337
0.059
5.661
0.000
H6
Sa is ac ion → Pu chase In en ion
0.291
0.294
0.067
4.343
0.000
H7
AI-Pe sonaliza ion → T us → Pu chase
In en ion
0.192
0.196
0.052
3.692
0.000
H8
AI-Pe sonaliza ion → T us → Sa is ac ion
0.165
0.168
0.051
3.235
0.001
H9
AI-Pe sonaliza ion → Engagemen → Pu chase
In en ion
0.207
0.211
0.055
3.764
0.000
H10
AI-Pe sonaliza ion → Sa is ac ion → Pu chase
In en ion
0.228
0.232
0.054
4.222
0.000
H11
P i acy Awa eness × AI-Pe sonaliza ion →
Engagemen
-0.062
-0.061
0.037
1.676
0.094
H12
P i acy Awa eness × AI-Pe sonaliza ion →
Sa is ac ion
-0.074
-0.073
0.036
2.056
0.040
Copy igh © ISRG Publishe s. All igh s Rese ed.
DOI: 10.5281/zenodo.17330694
275
H13
P i acy Awa eness × AI-Pe sonaliza ion →
Pu chase In .
-0.089
-0.087
0.039
2.282
0.023
H14
P i acy Awa eness × (AI-Pe sonaliza ion →
T us → Pu chase In .
-0.071
-0.072
0.034
2.088
0.037
5. Discussion
The indings o his s udy p o ide s ong e idence ha AI-powe ed
pe sonaliza ion signi ican ly shapes consume beha io ou comes.
The di ec pa hs con i med ha pe sonaliza ion enhances
engagemen , sa is ac ion, and pu chase in en ion (H1–H3). These
esul s demons a e ha when consume s ecei e ailo ed
ecommenda ions and ele an con en , hey a e mo e likely o
become in ol ed wi h a b and, expe ience g ea e enjoymen in he
in e ac ion, and ul ima ely de elop s onge pu chase in en ions
[37]. This highligh s he cen al ole o pe sonaliza ion as no only
a ma ke ing ac ic bu also a d i e o consume empowe men in
ou ism, whe e a ele s eel ecognized and alued h ough
ailo ed expe iences. Fo example, online a el agencies and
booking pla o ms inc easingly use pe sonaliza ion engines o
ecommend des ina ions, ho els, and ac i i ies based on p io
sea ches, pas ips, and s a ed p e e ences [38]. By aligning o e s
wi h indi idual a el in e es s—such as ad en u e ou s, cul u al
a ac ions, o luxu y s ays— hese pla o ms s eng hen cus ome
engagemen and loyal y while inc easing booking in en ions.
Consume us eme ged as a c i ical de e minan o pu chase
in en ion (H4). The esul s show ha us s eng hens he link
be ween pe sonaliza ion and consume ou comes, bo h di ec ly and
as a media o (H7–H8). This unde sco es ha pe sonaliza ion is
mos e ec i e when accompanied by pe cep ions o secu i y,
ai ness, and esponsible handling o a ele da a. Wi hou us ,
he bene i s o pe sonaliza ion may no ansla e in o conc e e
beha io al ou comes such as sa is ac ion o booking in en ion. A
good example is he ou ism and hospi ali y sec o , whe e AI-
d i en pe sonalized a el ecommenda ions, dynamic p icing, o
ailo ed holiday packages a e only e ec i e when a ele s us
how hei pe sonal and loca ion da a a e being used. Wi hou ha
us , ou is s may ejec e en he mos ele an and a ac i e
a el o e s, pe cei ing hem ins ead as in usi e o manipula i e
[39].
The indings also alida e he in e dependency be ween
engagemen and sa is ac ion. Engagemen was shown o imp o e
sa is ac ion (H5), while sa is ac ion s ongly in luenced pu chase
in en ion (H6). This indica es ha engagemen unc ions as a
p ecu so o deepe posi i e consume e alua ions. Mo eo e , bo h
engagemen and sa is ac ion se ed as media o s (H9–H10),
ein o cing he idea ha pe sonaliza ion does no in luence booking
beha io in isola ion bu a he ope a es h ough hese ela ional
mechanisms. Fo ins ance, online a el pla o ms and ou ism
agencies engage a ele s wi h in e ac i e ecommenda ion
sys ems ha highligh des ina ions, ou s, and accommoda ions
ailo ed o indi idual in e es s. When hese sugges ions align wi h a
a ele ’s p e e ences—such as cul u al excu sions, amily- iendly
packages, o luxu y expe iences— hey no only cap u e a en ion
bu also enhance sa is ac ion, which in u n inc eases he likelihood
o booking [40,41].
The mos dis inc i e con ibu ion o his s udy lies in he analysis
o da a p i acy awa eness as a mode a ing ac o . The esul s
showed ha p i acy awa eness weakens he posi i e e ec o
pe sonaliza ion on sa is ac ion and pu chase in en ion (H12–H13),
as well as he indi ec pa h h ough us (H14). This inding
highligh s a ension in AI-d i en pe sonaliza ion: while
pe sonaliza ion c ea es alue by ailo ing expe iences, i s impac
diminishes when consume s a e highly awa e o p i acy issues.
Highly p i acy-conscious consume s may pe cei e pe sonaliza ion
no as help ul, bu as in usi e o manipula i e [9]. This
phenomenon can be obse ed in he e ail sec o , whe e highly
a ge ed ads some imes gene a e skep icism and ―ad a igue,‖
pa icula ly when consume s suspec ha hei b owsing his o y is
being acked wi hou consen . In e es ingly, he mode a ion o
p i acy awa eness on engagemen (H11) was no signi ican ,
sugges ing ha e en p i acy-conscious consume s may s ill
pa icipa e in pe sonalized expe iences a an ini ial le el o
cu iosi y o con enience. Fo example, consume s migh s ill click
on pe sonalized ashion o a el ecommenda ions, bu hei
longe - e m sa is ac ion o pu chase decisions will depend on
whe he p i acy conce ns a e add essed [8].
Compa ing hese indings wi h he b oade li e a u e, a clea dual
pe spec i e eme ges. On one side, pe sonaliza ion is celeb a ed o
educing cogni i e e o , enhancing con enience, and inc easing
he pe cei ed alue o o e s, he eby suppo ing s onge
consume –b and ela ionships [42]. On he o he side, esea ch
inc easingly poin s o he isks o o e - a ge ing, p i acy in asion,
and he e osion o au onomy, all o which can gene a e consume
esis ance [43]. This s udy suppo s bo h pe spec i es
simul aneously: pe sonaliza ion is powe ul, bu i s bene i s a e
condi ional. I is no a uni e sal posi i e o ce; a he , i s e ec s
depend hea ily on he psychological and e hical amewo ks wi hin
which consume s in e p e i .
Toge he , hese esul s emphasize ha AI pe sonaliza ion is a
double-edged swo d. On one side, i os e s engagemen ,
sa is ac ion, us , and pu chase in en ion, making i a aluable
s a egy o o ganiza ions seeking o enhance consume
ela ionships. On he o he , i s e ec i eness is cons ained by
consume s’ p i acy awa eness, which condi ions how
pe sonaliza ion is pe cei ed. These indings en ich exis ing deba es
in ma ke ing and consume esea ch by showing ha he u u e o
pe sonaliza ion depends no only on echnological ad ancemen
bu also on social accep ance [44]. Theo e ical con ibu ions o his
s udy he e o e lie in in eg a ing p i acy awa eness in o
pe sonaliza ion esea ch as a bounda y condi ion, while p ac ical
implica ions poin o he need o i ms o pu sue esponsible
pe sonaliza ion s a egies. O ganiza ions mus design
pe sonaliza ion ini ia i es ha no only deli e ele ance bu also
demons a e anspa ency and accoun abili y in da a use. In doing
so, businesses can esol e he ension be ween pe sonaliza ion and
p i acy, unlocking he bene i s o AI while sa egua ding consume
us .
5.1 Theo e ical Implica ions
The indings o his s udy con ibu e o heo y by ex ending he
unde s anding o AI-powe ed pe sonaliza ion beyond i s di ec
in luence on consume beha io . While p io wo k has la gely
ocused on sa is ac ion, engagemen , and us as ou comes [3,32],
his esea ch highligh s he impo ance o da a p i acy awa eness
as a bounda y condi ion ha shapes he e ec i eness o
