Using AI for mental health support: benefit vs limitations

Sometimes students turn to AI for mental health support but this is NOT a replacement for professional treatment, it is NOT intended for emergencies, and NOT for therapy.  Using AI for mental health is not without risks (noted below). A recent article discusses considerations for using Google Gemini AI for mental health (1).

Combining traditional wellness practices with digital tools like Google Gemini can provide more accessible, personalized support.

Here are some examples (1):

Step Action How Google Gemini Helps
1 Assessment & Baseline: Reflect on current emotional wellbeing habits, stress levels, sleep, etc. Gemini can help you create baseline surveys, interpret wearable data.
2 Set Goals: Specific, measurable, realistic mental health goals (e.g. reduce anxiety, improve sleep). Gemini can suggest goal‑setting frameworks, help you refine goals.
3 Plan Interventions: Choose from the practices above that suit you. Gemini can help pick appropriate interventions; schedule reminders.
4 Tools & Resources: Apps, guided meditations, wearable trackers. Gemini can help you identify additional resources that may be helpful.
5 Monitor & Iterate: Track your progress; note what works, what doesn’t. Gemini can analyze your logs, suggest adjustments.
6 Support Network & Professional Help: Use community,  professional therapy, peer support when needed. Gemini can help you locate local professionals, support groups, create checklists for sessions.  For mental health support options at OSU, go here: https://ccs.osu.edu/services/mental-health-resources
 

 

Using Google Gemini to enhance 10 Evidence‑Based Practices that support Mental health (1)

1 Mindfulness and Meditation such as seated meditation, body scan, mindful breathing.

  • Benefits: Reduced stress and anxiety, improved emotional regulation, increased attention span.
  • AI / Google Gemini’s role: Can generate personalized guided meditations; suggest mindfulness prompts; help analyze meditation logs; recommend apps, practices based on user’s mood.

2 Physical activity such as Aerobic exercise, strength training, yoga, etc.

  • Benefits: Releases endorphins; improves mood; reduces symptoms of depression and anxiety.
  • How to do it: Regular routine (e.g., 30 mins, 3‑5 times/week); choose types you enjoy.
  • AI / Google Gemini’s role: Reminders, custom workout plans; tracking progress; motivating messages; adapting plan based on feedback.

3 Adequate Sleep Hygiene such as consistent schedule, avoiding caffeine at night, limiting screen time before bed.

  • Benefits: Better mood, improved cognitive function, reduced risk of mental health disorders.
  • How to do it: Set regular wake/sleep times, create sleep‑friendly bedroom, avoid blue light at night.
  • AI / Google Gemini’s role: Suggest improvements; analyze sleep trackers; recommend sleep routines; issue alerts when patterns deteriorate.

4 Balanced Nutrition with whole foods: vegetables, fruits, lean proteins, healthy fats; reducing processed foods.

  • Benefits: Affects brain health (neurotransmitters); energy stability; mood stabilization.
  • How to do it: Meal planning; nutrition; hydration.
  • AI / Google Gemini’s role: Suggest recipes; smooth meal planning; help adjust nutrition to lifestyle; track nutritional deficiencies.

5 Social Connection such as Maintaining friendships, community engagement, supportive relationships.

  • Benefits: Lower rates of depression and anxiety; buffer stress; improve wellbeing.
  • How to do it: Regular catch‑ups; joining interest groups; volunteering; quality time.
  • AI / Google Gemini’s role: Reminders to reach out; suggest local groups; help draft messages; coach on communication skills.

6 Cognitive Behavioral Techniques such as reframing negative thoughts, behavioral activation.

  • Benefits: Strong evidence in reducing depression, anxiety; improving resilience.
  • How to do it: Identify negative thinking patterns; set small behavioral goals.
  • AI / Google Gemini’s role: Guide journaling; challenge unhelpful thoughts; suggest techniques.

7 Journaling and Reflective Practices such as Writing down thoughts, gratitude journaling, reflection on daily experience.

  • Benefits: Helps process emotions; increases self‑awareness; reduces rumination.
  • AI / Google Gemini’s role: Provide prompts; analyze themes over time; offer feedback; suggest reflection questions.

8 Limiting Screen Time & Digital Detox, especially social media or negative content; periodic breaks.

  • Benefits: Improves sleep, reduces anxiety, improves concentration.
  • How to do it: Set screen‑free hours; remove apps; use blue‑light filters; substitute with offline activities.
  • AI / Google Gemini’s role: Monitor usage; suggest schedule for detox; send reminders; provide alternative offline ideas.

9 Nature Exposure such as Time outdoors, green spaces, forests, parks, natural light.

  • Benefits: Reduces stress; improves mood; improves attention; sometimes lowers blood pressure.
  • How to do it: Daily walks; gardening; sitting outside; weekend hikes.
  • AI / Google Gemini’s role: Suggest nearby parks; remind to get outside; provide information on nature therapy; support tracking nature exposure.

10 Professional Support & Therapy: Talking to mental health professionals (therapists, psychologists, psychiatrists), possibly medication if needed.

  • Benefits: Tailored treatment; long‑term improvement; skills development.
  • How to do it: Seek licensed professional; assess online therapy options; ensure credentials; set expectations.
  • AI / Google Gemini’s role: Provide information on finding providers; clarify what therapy entails; prepare questions; help understand treatment options; supplement (not replace) professional help. https://ccs.osu.edu/services/mental-health-resources

Potential Risks and Ethical Considerations (1)

AI‑assisted mental wellness has promise, but also comes with risks, so being aware can help in using such tools safely.

  • Accuracy & Hallucinations: As studies show, models including Gemini may sometimes produce incorrect or misleading outputs. For medical or mental health matters, this can be harmful.
  • Privacy & Data Security: Mental health data, sensor data, journal entries are highly sensitive. Ensuring secure storage, consent, transparency in use is crucial. Understand terms and conditions and avoid entering private, confidential, individually identifiable information whenever possible.
  • Overreliance on AI / Self‑Diagnosis: Tools should support, not replace, professional help. Self‑managing with AI alone might delay getting necessary care.
  • Bias and Culture: Mental health concepts and practices are culture‑sensitive. What works in one region might not be valid in another. AI trained on biased datasets may misinterpret nonwestern expressions of distress.
  • Ethical / Regulatory Compliance: Data protection laws (e.g. HIPAA, GDPR), professional guidelines, licensing issues for digital health tools must be respected. Take time to familiarize your self with these features/ data limitations of the AI you are using.
  • Limitations: Limitations include potential inaccuracies, lack of emotional nuance, data privacy concerns, and an inability to provide licensed therapeutic interventions.

Additional resources: For mental health support options at OSU, go here: https://ccs.osu.edu/services/mental-health-resources

By Ryan S Patel DO, FAPA
OSU-CCS Psychiatrist
Contact: patel.2350@osu.edu

Disclaimer: This article is intended to be informative only. It is advised that you check with your own physician/mental health provider before implementing any changes.  With this article, the author is not rendering medical advice, nor diagnosing, prescribing, or treating any condition, or injury; and therefore claims no responsibility to any person or entity for any liability, loss, or injury caused directly or indirectly as a result of the use, application, or interpretation of the material presented.

References:

  1. https://www.quickobook.com/healthfeed/view/how-google-gemini-is-transforming-mental-wellness-10-proven-ways-to-improve-your-mind-and-mood

Can a Ketogenic Eating Pattern Help With Depression in College Students?

Depression is among one of the top mental health concerns among college students (1).  Today, studies suggest that nearly 50% of college students are now navigating some form of depression(2).

We conducted a study with OSU’s Dr. Jeff Volek from the Volek Low-Carbohydrate Laboratory in conducting the KIND (Ketogenic Intervention in Depression) study: a pilot evaluating the potential of a well-formulated ketogenic diet as an adjunctive therapy for college students living with major depressive disorder (3).

Here is what you need to know about a well-formulated ketogenic eating pattern:

  • This dietary intervention was originally developed at the Mayo Clinic in 1921 as a treatment for pediatric epilepsy and now has over a century of safety and efficacy data supporting its use in this population.
  • Numerous studies show the potential benefit of well-formulated ketogenic diets for a variety of physical health conditions. (4), with more studies ongoing.
  • Key studies from Stanford School of Medicine and Edinburgh University highlight the positive psychiatric effects of the ketogenic diet, supporting its potential as an exciting therapeutic option for various mental health conditions(5, 6).
  • A well-formulated ketogenic eating pattern (WFKD) is a nutritionist-guided, personalized eating pattern that uses whole foods such as fruits, vegetables, healthful fats, proteins and minerals in a manner that is low-carbohydrate and prioritizes healthful fats and proteins, which promotes ketone production. There is no calorie restriction. These ketones offer the brain an alternative energy source to glucose, and emerging research indicates this switch positively impacts both metabolic and mental health, providing new hope for people living with serious mental illness.

What was the study? (3)

  • We conducted a pilot study in 24 college students aged 18-30, diagnosed with major depressive disorder, BMI >/=20, and no eating disorder, pregnancy, or other contraindications, who are currently receiving treatment in the form of talk therapy and or medication (3).
  • The goal of the study was to assess whether college students with major depressive disorder could adhere to a nutritionist-guided, well-formulated ketogenic diet and what impact the diet could have on their depression symptoms and quality of life.
  • After baseline testing for mental and general health, participants were assisted in adopting a well-formulated ketogenic diet for 10-12 weeks. Nutritionist support was available throughout the study on an as-needed basis.

What were the results? (3)

  • WFKD was feasible: 

Most participants completed the study (67%)

7 out of 8 participants who dropped out did so for reasons unrelated to the diet

Participants were in ketosis the majority of the time (73%)

  • All participants who completed the study saw clinically significant improvements in depression as measured using validated instruments such as the Patient Health Questionnaire 9 (PHQ) and the Hamilton Depression Rating Scale (HDRS).
  • Participants experienced rapid, significant improvements in depressive symptoms:

2 weeks: 37% improvement (PHQ-9), near 2x increase in global wellness

6 weeks: 59% improvement (HDRS)

  • Final improvements in depression were robust:

69% improvement (PHQ-9)

71% improvement (HDRS)

Nearly a 3x improvement in WHO-5, a measure of global wellness 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

While the results are impressive, even more inspiring were the individual stories from participants in the study:

  • One participant shared: “After the second week, my depression symptoms were gone. They were gone” 
  • Another participant reported: “I felt great about myself, and I noticed that I would never really slip into those dark thoughts before I started the keto study.” 
  • Remarkably, some participants who were strong candidates for medication based on symptom severity before the study showed such improvements that they no longer met the criteria for medication by the end of the study.

What are some caveats and precautions?

  • This was a small study. To confidently apply these findings to a broader population, larger randomized controlled trials comparing this intervention to a control diet are needed.
  • Participants followed what is referred to as a “well-formulated ketogenic diet” consisting primarily of whole foods, unlike other mainstream versions of ketogenic diet, which may not yield such strong outcomes.
  • It is always best to implement a new mental health strategy under the care of a licensed healthcare provider.

What does this mean for you?

  • If you’re struggling with depression or even anxiety, it’s important to consider the impact what you do and don’t eat has on your mental well-being.
  • Our findings suggest that, for appropriately selected patients (See inclusion and exclusion criteria), a well-formulated ketogenic eating pattern can be a low-risk intervention that may complement existing mental health strategies to provide greater depression symptom relief, improve quality of life, and improve brain function.

The KIND study was funded by Baszucki Group, a non-profit aiming to deliver fundamental improvements in metabolic and mental health. Metabolic Mind is an education and resource platform created by the Baszucki family after their son recovered from a five-year battle with bipolar disorder using a therapeutic ketogenic diet. 

Want more strategies to support your mental health?

Visit Metabolic Mind’s website to learn more about metabolic psychiatry and well-formulated ketogenic eating pattern. Looking to get started with ketogenic therapy? Check out THINK+SMART, a free community-inspired framework that helps individuals employ metabolic therapies to improve mental health. 

 

By Ryan S Patel DO, FAPA
OSU-CCS Psychiatrist
Contact: patel.2350@osu.edu

Disclaimer: This article is intended to be informative only. It is advised that you check with your own physician/mental health provider before implementing any changes.  With this article, the author is not rendering medical advice, nor diagnosing, prescribing, or treating any condition, or injury; and therefore claims no responsibility to any person or entity for any liability, loss, or injury caused directly or indirectly as a result of the use, application, or interpretation of the material presented.

References:

  1. American College Health Association. American College Health Association-National College Health Assessment III: Reference Group Executive Summary Spring 2025. Silver Spring, MD: American College Health Association; 2025.
  2. Luo MM, Hao M, Li XH, Liao J, Wu CM, Wang Q. Prevalence of depressive tendencies among college students and the influence of attributional styles on depressive tendencies in the post-pandemic era. Front Public Health. 2024 Jan 25;12:1326582. doi: 10.3389/fpubh.2024.1326582. PMID: 38333740; PMCID: PMC10850216.
  3. Decker, D.D., Patel, R., Cheavens, J. et al. A pilot study examining a ketogenic diet as an adjunct therapy in college students with major depressive disorder. Transl Psychiatry 15, 322 (2025). https://doi.org/10.1038/s41398-025-03544-8
  4. Masood W, Annamaraju P, Khan Suheb MZ, et al. Ketogenic Diet. [Updated 2023 Jun 16]. In: StatPearls [Internet]. Treasure Island (FL): StatPearls Publishing; 2025 Jan-. Available from: https://www.ncbi.nlm.nih.gov/books/NBK499830/
  5. Shebani Sethi, Diane Wakeham, Terence Ketter, Farnaz Hooshmand, Julia Bjornstad, Blair Richards, Eric Westman, Ronald M Krauss, Laura Saslow, Ketogenic Diet Intervention on Metabolic and Psychiatric Health in Bipolar and Schizophrenia: A Pilot Trial, Psychiatry Research,Volume 335, 2024, 115866, ISSN 0165-1781, https://doi.org/10.1016/j.psychres.2024.115866. (https://www.sciencedirect.com/science/article/pii/S0165178124001513)
  6. Campbell IH, Needham N, Grossi H, et al. A pilot study of a ketogenic diet in bipolar disorder: clinical, metabolic and magnetic resonance spectroscopy findings. BJPsych Open. 2025;11(2):e34. doi:10.1192/bjo.2024.841

Risks of Students using AI for mental health

Artificial intelligence (AI) is becoming increasingly common and by some estimates, AI apps are one of the most popular apps in the world (1). Globally, nearly 700 million people accessed AI-centric apps, especially chatbots or image editing tools, in 2024 (2).

A nationwide survey reported that over 50% of students have used major AI platforms like ChatGPT or similar large language models for mental health advice, emotional support, or therapeutic conversations (3, 4).

What are some risks of using AI for mental health support?

There are media reports both on benefits and harms of using both general purpose and mental health specific AI.

The research results are mixed:

  • One review of 18 randomized controlled trials found that  AI based therapy chatbots programmed to use specific types of therapy only may reduce symptoms of anxiety and depression shows promising results (5). The study populations were limited, chatbot design and psychotherapeutic approaches varied among the studies; all of which may limit the generalizability (5).
  • A recent Stanford study found significant risks with AI therapy chatbots (6):
    • LLMs expressed stigma toward people with mental health conditions (6)
    • Failed to respond safely to suicidal ideation 20-50% of the time (compared to 93% appropriateness from human therapists) (6)
    • Could not form genuine therapeutic relationships, which are key predictors of therapy success (6)

What are some risks that students should be aware of when using AI for mental health?

  • Lack of Personalization: AI bots cannot fully understand trauma or human emotion, such that it is not human and do not have lived experiences, making them struggle to respond in the “correct” way. (7)
  • False sense of support: These apps might make college students avoid seeking professional help when necessary, which can have serious consequences for those who need the support.  (7)
  • Privacy concerns: AI companies may collect data that people input into the system, which raises the questions of who has access to your data and the condition of your mental health. (7)
  •  The JED Foundation and the American Psychological Association highlight the following risks (8,9):
    • Distorted reality and harmed trust. Generative AI (the type designed to complete tasks or convey information) and algorithmic amplification might spread misinformation, worsen body image issues, and enable realistic deepfakes, undermining young people’s sense of self, safety, and truth. (8,9)
    • Invisible manipulation. AI curates feeds, monitors behavior, and influences emotions in ways young people often cannot detect or fully understand, leaving them vulnerable to manipulation and exploitation. This includes algorithmic nudging and emotionally manipulative design. (8,9)
    • Content that can escalate crises. Reliance on chatbot therapy alone can be detrimental due to inadequate support and guidance. Due to the absence of clinical safeguards, chatbots and AI-generated search summaries may serve harmful content or fail to alert appropriate human support when someone is in distress, particularly for youth experiencing suicidal thoughts. (8,9)
    • Simulated support without care. Chatbots posing as friends or therapists may feel emotionally supportive, but they can reinforce emotional dependency, delay help-seeking, disrupt or replace real friendships, undermine relational growth, and simulate connection without care. This is particularly concerning for isolated or vulnerable youth who may not recognize the limits of artificial relationships. (8,9)
    • Deepening inequities. Many AI systems do not reflect the full variety of youth experience in a broad variety of populations. As a result, they risk reinforcing stereotypes, misidentifying emotional states, or excluding segments of the youth populatoin. (8,9)
  • Other considerations: (9)
    • AI programs may lack nuanced understanding of individual symptoms, ability to interpret/contextualize, and may have limited understanding of the individuals co-occurring conditions.
    • Be cautious of AI “sycophancy”
    • The programs are not perfect and there is potential for harmful advice: not to take at face value
    • There are risks of open ended questions to general ai’s for mental health
    • Is this usage displacing or augmenting human interactions?
    • Discontinue use if harmful or unhelpful
  • Finally, AI is not intended for emergencies or to replace professional treatment.
  • While there are some commercially available programs that use AI and may be beneficial for structured activities such as  sleep log, mood chart, learn to implement and practice personalized coping skills and techniques, to assist in connecting with healthy life behaviors, increase connection with others; research and development is ongoing and students should proceed with caution, keeping the risks in mind. Products, features, and safeguards are also evolving.

By Ryan S Patel DO, FAPA
OSU-CCS Psychiatrist
Contact: patel.2350@osu.edu

Disclaimer: This article is intended to be informative only. It is advised that you check with your own physician/mental health provider before implementing any changes.  With this article, the author is not rendering medical advice, nor diagnosing, prescribing, or treating any condition, or injury; and therefore claims no responsibility to any person or entity for any liability, loss, or injury caused directly or indirectly as a result of the use, application, or interpretation of the material presented.

References:

  1. https://backlinko.com/most-popular-apps
  2. https://www.businessofapps.com/data/ai-app-market/
  3. https://sentio.org/ai-blog/ai-survey
  4. Rousmaniere, T., Zhang, Y., Li, X., & Shah, S. (2025) Large Language Models as Mental Health Resources: Patterns of Use in the United States. Practice Innovations.
  5. Wenjun Zhong, Jianghua Luo, Hong Zhang. The therapeutic effectiveness of artificial intelligence-based chatbots in alleviation of depressive and anxiety symptoms in short-course treatments: A systematic review and meta-analysis.
    Journal of Affective Disorders. Volume 356,2024,Pages 459-469,ISSN 0165-0327,https://doi.org/10.1016/j.jad.2024.04.057. https://www.sciencedirect.com/science/article/pii/S016503272400661X
  6. Jared Moore, Declan Grabb, William Agnew, Kevin Klyman, Stevie Chancellor, Desmond C. Ong, and Nick Haber. 2025. Expressing stigma and inappropriate responses prevents LLMs from safely replacing mental health providers. In Proceedings of the 2025 ACM Conference on Fairness, Accountability, and Transparency (FAccT ’25). Association for Computing Machinery, New York, NY, USA, 599–627. https://doi.org/10.1145/3715275.3732039
  7. https://www.behavioralhealthtech.com/insights/benefits-and-risks-of-ai-for-college-students
  8. Tech Companies and Policymakers Must Safeguard Youth Mental Health in AI Technologies | The Jed Foundation.  https://jedfoundation.org/artificial-intelligence-youth-mental-health-pov/
  9. Health advisory: Artificial intelligence and adolescent well-being.  https://www.apa.org/topics/artificial-intelligence-machine-learning/health-advisory-ai-adolescent-well-being