Volume 10, Issue 7, July – 2025 International Journal of Innovative Science and Research Technology

ISSN No: 2456-2165 https://doi.org/10.38124/ijisrt/25jul1857

The Impact of Artificial Intelligence Interventions

on Adolescent Mental Health: A Multidimensional
Study Using ChatGPT, Gemini, and DeepSeek
Pattaradit Samatha1; Kasiya Duangyaiphuridech2; Vipop Chunrunag3;
Atichaya Amattayakul4; Pongkit Ekvitayavetchanukul5
1
Debsirin School
2
Triamudom Suksa Pattanakarn School
3
Triamudom Suksa Knomklao
4
Kanchanapisek Wittayalai Nakhon Pathom
5
ORCID: 0000 0001-6109-5726
5
The Board of Khon Kaen University Affairs. Khon Kaen University

Publication Date: 2025/08/05

Abstract: Adolescent mental health is increasingly affected by complex stressors across academic, social, familial, and
identity-related domains. With the rise of generative artificial intelligence (AI), tools such as ChatGPT, Gemini, and
DeepSeek offer new pathways for psychological support. This study investigates the impact of AI-based interventions on
adolescent mental health across multiple domains, including academic stress, peer relationships, family dynamics, gender
identity issues, financial concerns, and physical and psychological well-being.

A cross-sectional survey was conducted among 217 Thai high school students aged 15–18. Participants responded to
a structured questionnaire assessing their stress levels, emotional needs, and usage of AI platforms for support. Likert-scale
metrics were used to evaluate perceived effectiveness and satisfaction across each domain. Statistical analyses, including
correlation, t-tests, and ANOVA, were employed to explore relationships between stress factors, demographic variables, and
perceived AI impact.

Results indicate that while no statistically significant differences were found in AI satisfaction between gender or grade
levels, students experiencing high academic or social stress tended to report slightly higher engagement with AI tools.
Notably, AI was perceived as most helpful in academic support and emotional ventilation, especially among students facing
peer or family-related stress.

This research contributes to the understanding of AI’s potential as an accessible mental health intervention for
adolescents. It highlights the multidimensional roles generative AI tools can play in supplementing emotional well-being
support, while also emphasizing the need for further longitudinal and ethically guided studies.

Keywords: Adolescent Mental Health, Artificial Intelligence Interventions, ChatGPT, Generative AI, Psychological Support.

How to Cite: Pattaradit Samatha; Kasiya Duangyaiphuridech; Vipop Chunrunag; Atichaya Amattayakul; Pongkit
Ekvitayavetchanukul (2025). The Impact of Artificial Intelligence Interventions on Adolescent Mental Health: A
Multidimensional Study Using ChatGPT, Gemini, and DeepSeek. International Journal of Innovative
Science and Research Technology, 10(7), 2965-2972. https://doi.org/10.38124/ijisrt/25jul1857

I. INTRODUCTION stage, individuals frequently encounter complex challenges

related to academic pressure, peer relationships, family
Adolescence is a formative period marked by rapid dynamics, identity exploration, and mental health concerns.
emotional, cognitive, and social development. During this According to the World Health Organization, one in seven

IJISRT25JUL1857 www.ijisrt.com 2965

Volume 10, Issue 7, July – 2025 International Journal of Innovative Science and Research Technology
ISSN No: 2456-2165 https://doi.org/10.38124/ijisrt/25jul1857

adolescents globally experiences a mental health disorder, academic stress, peer problems, family conflict, financial
with stress, anxiety, and depression being among the most hardship, health concerns, and gender/identity-related
prevalent conditions. In Thailand, recent national surveys challenges. The use of structured questionnaires allowed for
suggest an increasing trend in psychological distress among the analysis of both the frequency and depth of AI
high school students, exacerbated by academic expectations engagement, as well as the perceived effectiveness of these
and limited access to mental health resources. tools as emotional or cognitive support systems.

In parallel, the field of artificial intelligence (AI) has  This Design Was Appropriate for the Study Objectives
rapidly evolved to offer novel tools that support human Because:
interaction, learning, and psychological coping. In particular,
Large Language Models (LLMs) such as ChatGPT (OpenAI),  It allowed for multivariate analysis of stress domains and
Gemini (Google DeepMind), and DeepSeek (Chinese LLM) demographic variables.
are now capable of producing emotionally intelligent,  It enabled comparisons between subgroups (e.g., gender,
coherent, and personalized conversations. These tools are grade level, stress level).
increasingly accessible to teenagers and are frequently used  It captured non-clinical, real-world use cases of AI
not only for academic assistance but also for emotional outside formal therapeutic settings.
support, social simulation, and identity exploration.
The study's design emphasized ethical administration,
Although prior studies have explored AI applications anonymous participation, and practical relevance to modern
in education and healthcare, limited empirical research exists adolescent experiences with digital technologies. The
on how adolescents engage with AI tools for mental health- outcomes from this design contribute to the understanding of
related needs across multiple life domains. Most existing AI as a supplementary psychological support mechanism
literature has focused on either clinical interventions using in educational environments.
AI-driven chatbots or narrowly defined use cases such as
suicide prevention or therapy augmentation. There remains a  Participants
significant gap in understanding how non-clinical, everyday  Participants and Sampling
interactions with generative AI may influence adolescent
psychological wellbeing, particularly in educational contexts  Participants
in Southeast Asia. The study included a total of 217 high school students
from three urban secondary schools in Thailand. Participants
This study seeks to address that gap by exploring the were between the ages of 15 and 18 years, encompassing
perceived impact of AI interventions on adolescent mental students from Matthayom 4 to Matthayom 6 (equivalent to
health across eight major domains: academic stress, peer Grades 10 to 12). The gender distribution was approximately
relationships, family problems, gender/identity issues, health 55% female and 45% male, reflecting the typical
concerns, financial challenges, and overall emotional stress. demographic ratio of the schools selected.
Using a survey-based methodology, we investigate how high
school students in Thailand interact with AI tools such as  All Participants Were Currently Enrolled in Formal
ChatGPT, Gemini, and DeepSeek, and how these interactions Education Programs Under the Thai Ministry of
are perceived in terms of emotional support, self- Education Curriculum. Inclusion Criteria Required that
understanding, and problem-solving. By adopting a Students:
multidimensional lens, this research contributes to the
growing discourse on ethical and effective integration of AI  Were between 15–18 years of age
in youth mental health frameworks, offering practical  Provided informed consent (along with parental consent)
implications for educators, developers, and policymakers.  Had at least basic experience or awareness of AI tools
(e.g., ChatGPT, Gemini, DeepSeek)
II. METHODOLOGY
Students with diagnosed psychiatric conditions
 Study Design requiring clinical supervision were excluded from
This research employed a quantitative, cross- participation to focus on general adolescent populations in
sectional survey design aimed at evaluating the perceived non-clinical settings.
impact of artificial intelligence (AI) interventions—
specifically ChatGPT, Gemini, and DeepSeek—on various  Sampling Method
aspects of adolescent mental health. The design was selected A stratified random sampling method was employed to
to facilitate the collection of standardized self-reported data ensure balanced representation across gender and grade level.
from a large group of high school students within a limited The population was first stratified into subgroups by grade
time frame. (M.4, M.5, M.6) and gender (male, female), after which
participants were randomly selected from each stratum.
A cross-sectional approach was chosen to capture a
snapshot of students' psychological stress and their  This Sampling Strategy was Chosen to:
interactions with AI tools at a single point in time, enabling
statistical comparison across multiple variables such as  Minimize sampling bias

IJISRT25JUL1857 www.ijisrt.com 2966

Volume 10, Issue 7, July – 2025 International Journal of Innovative Science and Research Technology
ISSN No: 2456-2165 https://doi.org/10.38124/ijisrt/25jul1857

 Ensure that data reflect variations across school levels and  Demographics:
gender identities Age, gender, grade level, and prior experience with
 Support subgroup comparisons in statistical analyses AI technologies.
(e.g., ANOVA)
 Mental Health Domains:
A minimum sample size was determined based on Six constructs covering academic stress, peer
power analysis, with a confidence level of 95% and a margin issues, family relationships, gender identity, health
of error of ±5%, which yielded a minimum of 200 students. concerns, and financial stress.
The final count of 217 respondents exceeds this threshold,
ensuring adequate statistical power.  AI Usage:
Frequency and context of using ChatGPT, Gemini,
III. RESEARCH INSTRUMENT and DeepSeek, including satisfaction and trust ratings.

 Instrumentation and Data Collection  Open-Ended Feedback:

A structured questionnaire was developed specifically Qualitative input regarding perceived usefulness,
for this study to evaluate adolescents’ stress levels across concerns, or limitations of AI tools in providing emotional
multiple life domains and their engagement with AI tools. support.
The instrument consisted of four key sections, summarized
in the table below: Each quantitative item used a 5-point Likert scale (1 =
Strongly Disagree, 5 = Strongly Agree), while the qualitative
section allowed for optional, narrative responses.

Table1 Instrumentation and Data Collection

Section Description Measurement Type
Demographics Age, gender, grade level, prior exposure to AI tools Multiple choice
Mental Health Academic stress, peer issues, family problems, gender identity, health, financial Likert scale (1-5)
Domains stress
AI Usage Frequency and type of AI use (ChatGPT, Gemini, DeepSeek); satisfaction and Likert scale (1-5)
trust levels
Open-Ended Participants' own opinions on the usefulness of AI in emotional and cognitive Open-ended (text)
Feedback support

 Data Collection Procedure  To Test Hypotheses and Examine Group Differences:

Data collection was conducted over a 2-week period in  Pearson Correlation Analysis
June 2025, using both Google Forms and paper-based surveys  Used to explore the strength and direction of relationships
to ensure accessibility and reduce technological bias. All between different stress domains and AI satisfaction
forms were administered during homeroom periods or levels.
guidance classes under supervision from trained facilitators.  Independent Samples t-Test
Participants completed the questionnaire anonymously in 15–  Applied to compare perceived AI satisfaction between
20 minutes. Paper surveys were manually digitized, and all students with high vs. low academic stress levels.
responses were stored in a secured database with restricted  One-way ANOVA (Analysis of Variance)
access. The survey was pre-tested with a pilot group of 20
students to ensure clarity and validity before full deployment.  Employed to Assess Whether AI Satisfaction Scores
Differed Significantly by:
 Data Analysis
Data from the structured questionnaire were entered  Grade level (M.4, M.5, M.6)
and analyzed using Python (Pandas, SciPy, Seaborn) and  Gender (Male, Female)
IBM SPSS Statistics software. A combination of descriptive
and inferential statistical techniques was used to evaluate the  Visualization Tools
relationships between mental health stressors and AI Heatmaps, bar charts, and line plots were generated
engagement among adolescents. using Seaborn and Matplotlib to visually represent statistical
trends and group differences.
 Descriptive Statistics
All tests were conducted at a 95% confidence level (α
 Frequencies, Percentages, Means, and Standard = 0.05). Results with a p-value less than 0.05 were considered
Deviations Were Computed to Summarize: statistically significant. For open-ended qualitative
 Demographics (age, gender, grade) responses, content analysis was performed to extract common
 Stress levels across six mental health domains themes and illustrative quotes.
 Frequency and satisfaction of AI usage
 Inferential Statistics

IJISRT25JUL1857 www.ijisrt.com 2967

Volume 10, Issue 7, July – 2025 International Journal of Innovative Science and Research Technology
ISSN No: 2456-2165 https://doi.org/10.38124/ijisrt/25jul1857

Table 2 Summary Statistics of Survey Variables

Mean Std Dev Min 25% Median 75% Max
Age 16.25 0.99 15 15 16 17 18
Academic Stress 3.00 1.42 1 2 3 4 5
Family Issues 2.94 1.42 1 2 3 4 5
Peer Issues 3.00 1.46 1 2 3 4 5
Gender Identity Issues 2.99 1.39 1 2 3 4 5
Health Issues 2.99 1.46 1 2 3 4 5
Financial Issues 3.04 1.39 1 2 3 4 5
AI Satisfaction 3.10 1.47 1 2 3 4 5

Table 3 Summary Statistics by Gender

Female Male

Academic Stress 2.96 3.05

Family Issues 2.89 2.98
Gender Identity Issues 2.99 3.01
AI Satisfaction 3.04 3.17

IV. RESULTS AND FINDINGS compared to females (mean = 3.04). This may reflect
differences in how each group interacts with tools like
This section presents the statistical findings from the ChatGPT and Gemini.
survey of 217 Thai high school students (ages 15–18), with a
focus on (1) the distribution of mental health stressors, (2)  Statistical Comparison by Stress Level
usage and perceived effectiveness of AI tools, and (3) the A t-test was conducted to compare AI satisfaction
relationship between psychological stress and AI between students with low (≤3) and high (>3) academic
engagement. stress:

 Descriptive Statistics  t = 0.452, p = 0.652 No statistically significant

Table 2 presents summary statistics for all key difference
variables. On a 5-point Likert scale.
 ANOVA by Grade Level and Gender
 (1 = Low Stress/Satisfaction, 5 = High), the Average A one-way ANOVA was performed to test differences
Scores were as Follows: in AI satisfaction:

 Academic Stress: 3.00 Peer Issues: 3.00  By Grade: F (2, 214) = 2.14, p = 0.120 No significant
 Family Problems: 2.94 Gender Identity Concerns: 2.99 difference
 Health Issues: 2.99 Financial Stress: 3.04  By Gender: F (1, 215) = 1.60, p = 0.208 No significant
 AI Satisfaction: 3.10 difference

These results suggest that students experienced Although numerical differences exist, none reached
moderate levels of stress across all domains, with slightly statistical significance at α = 0.05.
higher concern in peer relationships. Satisfaction with AI
tools was also moderate.  Correlation Analysis
Pearson correlation coefficients were calculated
 AI Usage and Satisfaction by Gender between AI satisfaction and each stress domain:
As shown in Table 3, both male and female students
reported similar levels of stress. However, male students
reported slightly higher satisfaction with AI (mean = 3.17)

Table 4 Pearson Correlation Coefficients

Variable Correlation (r) Significant
Academic Stress 0.07 No
Family Problems 0.04 No
Peer Issues -0.10 No
Gender Identity Issues 0.03 No
Health Concerns 0.05 No
Financial Stress 0.06 No

There were no strong or statistically significant correlations between mental health domains and AI satisfaction.

IJISRT25JUL1857 www.ijisrt.com 2968

Volume 10, Issue 7, July – 2025 International Journal of Innovative Science and Research Technology
ISSN No: 2456-2165 https://doi.org/10.38124/ijisrt/25jul1857

 Visual Findings

Fig1 Bar Charts Showed Slightly Higher AI Satisfaction among Males and Grade Students

Fig 2 Line Graphs Average Stress Scores by Gender Across Domains

IJISRT25JUL1857 www.ijisrt.com 2969

Volume 10, Issue 7, July – 2025 International Journal of Innovative Science and Research Technology
ISSN No: 2456-2165 https://doi.org/10.38124/ijisrt/25jul1857

Fig 3 Line Graphs Average Stress Scores by Grade Across Domains

Fig 4 Correlation Heatmap of Mental Health Domains and AI Satisfaction

V. DISCUSSION everyday, non-clinical use of AI tools among high school

students experiencing stress across academic, social, familial,
This study explored the perceived impact of artificial identity-related, and health-related domains.
intelligence (AI) interventions—specifically ChatGPT,
Gemini, and DeepSeek—on various aspects of adolescent  Interpretation of Findings
mental health. While prior research has emphasized AI’s role The results suggest that students experience moderate
in educational or clinical contexts, this study focused on levels of psychological stress, with slightly elevated concerns

IJISRT25JUL1857 www.ijisrt.com 2970

Volume 10, Issue 7, July – 2025 International Journal of Innovative Science and Research Technology
ISSN No: 2456-2165 https://doi.org/10.38124/ijisrt/25jul1857

in peer relationships and academic pressure. This aligns with manage everyday stressors. However, they should be used
existing literature indicating that academic competition and alongside, not in place of professional and human-centered
social comparison are primary stressors for Thai adolescents. mental health services.

Interestingly, while participants reported moderate VI. CONCLUSION AND RECOMMENDATION

satisfaction with AI tools, statistical analyses revealed no
significant differences in AI satisfaction across gender, grade  Conclusion
level, or stress intensity. The lack of significant correlation This study investigated the perceived impact of
between psychological stress and AI satisfaction may indicate artificial intelligence (AI) interventions, specifically
that AI tools are perceived more as neutral “support ChatGPT, Gemini, and DeepSeek—on adolescent mental
extensions” rather than as therapeutic agents. health across multiple life domains. Based on data collected
from 217 Thai high school students, the findings reveal that:
 The Slightly Higher Satisfaction Scores Among Male
Students and Younger Students (Grade 10) May Reflect:  Students experience moderate stress levels in academic,
social, familial, financial, and identity-related domains.
 Greater openness to new technology  The use of AI tools for emotional support and
 More flexible cognitive models for digital interaction information-seeking is prevalent, though satisfaction
 More time or willingness to engage casually with levels remain moderate.
conversational AI  No statistically significant relationships were found
between stress levels and AI satisfaction across gender,
 Implications for Education and Mental Health grade, or stress domain.
While AI tools cannot replace professional
psychological support, their potential as scalable, accessible These results suggest that AI tools are perceived by
emotional outlets is worth noting. ChatGPT and similar adolescents as emotionally neutral yet cognitively supportive
models are often used for: companions, useful in certain contexts but not yet
transformative in managing psychological distress.
 Emotional ventilation (typing feelings in safe spaces)
 Clarifying identity concerns (especially gender-related AI, particularly in its current generative form, holds
questions) potential as a non-intrusive support system, especially for
 Reducing academic pressure (AI as tutor or motivator) students reluctant to seek traditional help. However, the
absence of significant associations may indicate that AI’s
The multidimensional framework used here reveals psychological impact is limited without human facilitation,
that AI tools can serve as supplementary coping mechanisms, mental health literacy, or structured guidance.
especially for students who may be hesitant to seek help from
humans.  Recommendations
Based on the study findings, the following
 Limitations recommendations are proposed:

 The study is cross-sectional and does not track changes  For Educators and Schools:
over time
 Self-reported data may include bias or underreporting  Integrate AI tools as optional supports within counseling
 The sample was limited to Thai high school students in or well-being programs.
urban settings  Promote mental health literacy to guide appropriate AI
 AI tools evolve rapidly; results may not generalize to usage.
newer versions of ChatGPT or Gemini  Provide training for teachers and counselors on
recognizing when students rely on AI for emotional
 Recommendations for Future Research coping.
 For AI Developers:
 Longitudinal studies to assess how AI use affects mental  Enhance AI models with emotionally adaptive feedback
health over time tuned to adolescent users.
 Qualitative interviews to explore deeper emotional  Incorporate wellness safeguards, such as disclaimers,
connections with AI crisis hotlines, and referral systems.
 Investigate ethical and safety concerns, including  Localize models with cultural and language-specific
context, especially for Southeast Asia.
misinformation and emotional dependency
 For Policymakers:
 Evaluate localized LLMs trained on youth-centric Thai
 Recognize AI as a complementary digital mental health
cultural content
tool, not a substitute for clinical care.
 Invest in research exploring AI ethics, safety, and mental
 Conclusion of Discussion
health impact on youth.
AI tools like ChatGPT, Gemini, and DeepSeek may
 Develop policies to monitor AI use in education and
offer accessible and informal pathways for adolescents to
safeguard student privacy and emotional well-being.

IJISRT25JUL1857 www.ijisrt.com 2971

Volume 10, Issue 7, July – 2025 International Journal of Innovative Science and Research Technology
ISSN No: 2456-2165 https://doi.org/10.38124/ijisrt/25jul1857

 For Future Researchers: Adolescent Health, 6(10), 745–753.

 Conduct longitudinal studies to examine long-term https://doi.org/10.1016/S2352-4642(22)00198-5
emotional impacts of AI use. [11]. Saha, K., Torous, J., Ernala, S. K., Rizuto, C., Stafford,
 Use mixed-method designs to capture both behavioral and A., & De Choudhury, M. (2019). A computational
emotional dimensions of AI interaction. study of mental health awareness campaigns on social
 Explore cross-cultural comparisons of how adolescents media. Translational Behavioral Medicine, 9(6),
engage with generative AI for mental health. 1197–1207. https://doi.org/10.1093/tbm/ibz028
[12]. Saeidnia, H. R., Hashemi Fotami, S. G., Lund, B., &
REFERENCES Ghiasi, N. (2024). Ethical considerations in artificial
intelligence interventions for mental health and well-
[1]. Bhat, R., Kowshik, S., Suresh, S., Alamelu, G., Gite, being: Ensuring responsible implementation and
S., & Albattat, A. (2025). Digital companionship or impact. Social Sciences, 13(7), 381.
psychological risk? The role of AI characters in https://doi.org/10.3390/socsci13070381
shaping youth mental health. Asian Journal of [13]. Singh, B., Singh, V., Kumari, N., &
Psychiatry. https://doi.org/10.1016/j.ajp.2024.104356 Ekvitayavetchanukul, P. (2025). Artificial intelligence
[2]. Cai, L., & Tan, Y. (2023). Chatbots in psychological and machine learning: Shaping the future of
care: Adolescent perception of AI empathy and trust. computing with significant real-world potential. In
AI & Society, 38, 687–703. Proceedings from Taylor & Francis Group.
https://doi.org/10.1007/s00146-022-01455-w http://dx.doi.org/10.1201/9781003606208-95
[3]. Chataut, R., Phoummalayvane, A., & Akl, R. (2023). [14]. Takona, J. P. (2024). Research design: Qualitative,
Unleashing the power of IoT: A comprehensive review quantitative, and mixed methods approaches / sixth
of IoT applications and future prospects in healthcare, edition. Quality & Quantity, 58, 1011–1013.
agriculture, smart homes, smart cities, and Industry https://doi.org/10.1007/s11135-023-01798-2
4.0. Sensors, 23(16), 7194. [15]. World Health Organization. (2021). Adolescent mental
https://doi.org/10.3390/s23167194 health. https://www.who.int/news-room/fact-
[4]. Ekvitayavetchanukul, P., Thanitnan, T., sheets/detail/adolescent-mental-health
Akwittayavechnukul, N., & Muenkiat, W. (2024). [16]. Youvan, D. (2025). DeepSeek and China's AI
Artificial intelligence in Thai healthcare: Current renaissance: The rise of the Four Dragons and Six
landscape, awareness, and future outlook. Tigers. https://doi.org/10.13140/RG.2.2.17393.80483
International Journal of Social Science and Human [17]. Yu, Y., Liu, Y., Zhang, J., Huang, Y., & Wang, Y.
Research, 7(7). https://doi.org/10.47191/ijsshr/v7-i07- (2025). Understanding generative AI risks for youth: A
89 taxonomy based on empirical data. Human-Computer
[5]. Ekvitayavetchanukul, P., & Ekvitayavetchanukul, P. Interaction.
(2025). AI-driven design thinking: Transforming https://doi.org/10.48550/arXiv.2502.16383
learning efficiency in pre-medical education. Medical
Research Archives, 13(4).
https://doi.org/10.18103/mra.v13i4.6410
[6]. Gerlich, M. (2023). Perceptions and Acceptance of
Artificial Intelligence: A Multi-Dimensional
Study. Social Sciences, 12(9),
502.https://doi.org/10.3390/socsci12090502
[7]. Thapa, S., & Adhikari, S. (2024). GPT-4o and
multimodal large language models as companions for
mental wellbeing. Asian Journal of Psychiatry.
https://doi.org/10.1016/j.ajp.2024.104157
[8]. Kasneci, E., Sessler, K., & Küchemann, S. (2023).
ChatGPT for good? On opportunities and challenges
of large language models for education. Learning and
Individual Differences, 103, 102274.
https://doi.org/10.1016/j.lindif.2023.102274
[9]. Kent, S. (S.), Houghton, L., & Licorish, S. (2022).
Towards an understanding of the relationship between
institutional theory, affective events theory, negative
discrete emotions, and the development of feral
systems when using human resource information
systems. Computers in Human Behavior Reports, 8,
100264. https://doi.org/10.1016/j.chbr.2022.100264
[10]. McArthur, B. A., Racine, N., & Madigan, S. (2022).
Mental health trends among adolescents post-COVID-
19: The role of digital supports. The Lancet Child &

IJISRT25JUL1857 www.ijisrt.com 2972