Paubox blog: HIPAA compliant email - easy setup, no portals or passcodes

How ChatGPT is transforming telemedicine

Written by Tshedimoso Makhene | August 30, 2025

Telemedicine has revolutionized healthcare by making medical services more accessible, efficient, and convenient. What began as a niche solution to connect patients in remote areas with healthcare professionals has now become a mainstream mode of care delivery. According to a 2021 8-week AMA survey, the AMA found that “85% of physician respondents indicate they currently use telehealth.” Furthermore, the survey found that 69% of the physicians were interested in continuing to provide telehealth services beyond the COVID-19 pandemic.

As Brian William Hasselfeld, M.D., assistant medical director for digital health innovations at Johns Hopkins Medicine, states, “Telemedicine, which enables video or phone appointments between a patient and their health care practitioner, benefits both health and convenience. More health care providers are offering to “see” patients by computer and smartphone… Improved technology has made telemedicine easier, even for those who don’t consider themselves computer savvy.”

This shift has redefined the patient experience. Instead of long commutes, crowded waiting rooms, or delayed appointments, patients can now connect with a provider in minutes using a smartphone or laptop. For providers, telemedicine enables broader reach, continuity of care, and new efficiencies in managing chronic conditions. Yet telemedicine is no longer just about the video call. The integration of artificial intelligence (AI), particularly conversational AI like ChatGPT, is taking this transformation even further.

By enabling natural, human-like dialogue, ChatGPT can act as both a support system for providers and a guide for patients, streamlining communication, improving health literacy, and reducing administrative burdens.

 

How and why is ChatGPT relevant to telemedicine?

ChatGPT is an AI-driven conversational model that simulates natural dialogue. Unlike traditional chatbots, which rely on fixed scripts, ChatGPT can adapt to the context of a conversation, making its interactions more dynamic and helpful. As stated in its statement at the release of GPT-4, “GPT‑4 is a large multimodal model (accepting image and text inputs, emitting text outputs) that, while less capable than humans in many real-world scenarios, exhibits human-level performance on various professional and academic benchmarks.”

As stated in the study Integrating ChatGPT-like models in telehealth portals, AI models, including ChatGPT, “Serve as virtual assistants, improving patient engagement and access.” While it does not replace medical expertise, it can complement telehealth systems in several important ways:

  • Patient-facing interactions: answering common questions, symptom guidance, educational support.
  • Provider-facing assistance: summarizing consultations, generating draft notes, supporting clinical decision-making.
  • Operational support: appointment scheduling, follow-ups, and billing inquiries.

See also: HIPAA Compliant Email: The Definitive Guide (2025 Update)

 

Applications of ChatGPT in telemedicine

According to the study Integrating ChatGPT-like models in telehealth portals, ChatGPT-like models are showing strong potential across clinical, educational, and administrative areas. By enabling natural, context-aware conversations, the technology can help address common challenges in digital healthcare. The study notes some key applications where ChatGPT is making the biggest impact, from improving patient communication to streamlining workflows.

 

Remote patient engagement and triage

ChatGPT-like models facilitate symptom assessment, guiding patients through conversational interfaces to determine urgency and direct them to appropriate care (e.g., self-care advice or emergency services). The authors note that “Their ability to process unstructured inputs allows flexibility in patient descriptions, making them particularly useful in preliminary diagnostics.  These  systems  can  also  suggest  self-care options  or  direct  users  to emergency services when needed.”

 

Medication guidance and adherence support

These systems help patients understand dosage instructions, potential side effects, and drug interactions. “ChatGPT-like systems can explain drug interactions, provide reminders,  and even monitor adherence patterns by querying patients about missed doses. This is especially useful for chronic disease management, where medication adherence is crucial.” They can also monitor adherence, potentially linking with digital pillboxes or wearable devices to send reminders.

 

Mental health support and companionship

ChatGPT-based systems are being explored to provide low-intensity mental health assistance, such as stress-relief prompts, journaling support, mood tracking, and cognitive behavioral strategies. They offer a 24/7 anonymous companion, especially useful for those hesitant to seek traditional care. However, the study also notes that “these  applications  must  be  carefully  designed  to  avoid triggering or harmful responses and should include escalation paths for severe cases.”

Related: How can mental health professionals benefit from ChatGPT?

 

Administrative workflow support

ChatGPT can be embedded to manage administrative tasks, such as scheduling appointments, sending reminders, managing follow-ups, and generating post-visit summaries, thereby reducing clinician workload and enhancing patient engagement. The authors of the study note that “Seamless integration of ChatGPT-like models into telehealth platforms requires thoughtful user interface design. These AI tools should be embedded in ways that complement clinical workflows and enhance user experience. The interface must be accessible across devices, support dynamic interactions, and offer visual cues for clarity. Personalization options such as tone adjustment or language preference can further enhance usability. The goal is to ensure that AI does not disrupt the user journey but instead adds value through clarity and convenience.”

 

Integration and accessibility features

When integrated into telehealth platforms, ChatGPT can appear within patient dashboards, pre-consultation forms, or post-visit sections. Options such as multilingual support, speech-to-text (STT), text-to-speech (TTS), and accessibility features for neurodivergent or elderly users make the experience inclusive.

 

Behind-the-scenes integration

The model’s deployment integrates via secure APIs with systems like EHRs, calendar tools, and CRM platforms. It supports auditing, logging, and responsiveness to clinical context while complying with regulations like the Health Insurance Portability and Accountability Act (HIPAA) or the General Data Protection Regulation (GDPR).

Read also

 

Ethical, legal, and regulatory considerations

Despite the clear benefits, integrating ChatGPT into telemedicine raises important concerns.

  • Accuracy and reliability: ChatGPT is not a medical professional. While it can provide helpful information, incorrect or misleading responses could pose risks. For example, if a patient asks about chest pain and the AI downplays it, the consequences could be life-threatening.
  • Privacy and security:Telemedicine platforms must comply with regulations such as HIPAA and GDPR. If ChatGPT is used to process sensitive patient data, developers must ensure encryption, access controls, and minimal data retention.
  • Bias and equity: AI models are trained on existing datasets, which may contain biases. If left unchecked, ChatGPT could unintentionally disadvantage certain demographic groups in healthcare advice.
  • Transparency: Patients must be informed when they are interacting with an AI tool rather than a human. This builds trust and ensures ethical use.
  • Liability: A critical question is: Who is responsible if ChatGPT provides harmful advice? Clear guidelines are needed to protect both patients and providers.

Read also: What risks does ChatGPT pose in a healthcare setting?

 

Using ChatGPT in a HIPAA compliant manner

While ChatGPT offers valuable applications in telemedicine, healthcare providers must take steps to ensure its use aligns with HIPAA requirements. Since ChatGPT itself is not inherently HIPAA compliant, the way it is deployed matters. Providers should:

  • Avoid entering protected health information (PHI) into publicly available versions of ChatGPT. PHI should only be shared if the AI is integrated into a HIPAA compliant platform under a signed business associate agreement (BAA).
  • Use secure, encrypted integrations where ChatGPT is hosted within a healthcare provider’s infrastructure or through vendors that offer HIPAA compliant AI solutions.
  • Apply strict access controls and data governance policies to prevent unauthorized disclosure of sensitive information.
  • Leverage ChatGPT for non-PHI tasks such as patient education materials, administrative support, and general health information guidance.

By implementing these safeguards, healthcare organizations can benefit from ChatGPT’s efficiency while protecting patient privacy and staying compliant with HIPAA standards.

Go deeper: A quick guide to using ChatGPT in a HIPAA compliant way

 

FAQS

Is ChatGPT HIPAA compliant?

No, ChatGPT is not HIPAA compliant. OpenAI does not currently enter into a business associate agreement (BAA) with healthcare organizations, which is a core requirement under HIPAA for any tool handling protected health information (PHI).

 

Is ChatGPT safe to use in healthcare?

ChatGPT can be safe if used responsibly and within HIPAA compliant systems. However, it should not be relied upon for diagnosis or treatment decisions without human oversight.