The rise of generative Artificial Intelligence (AI), spearheaded by companies like OpenAI with their groundbreaking programs such as ChatGPT and DALL-E, has captured global attention. While these tools showcase impressive capabilities in generating human-like text and diverse images, the potential of AI extends far beyond general applications, particularly into specialized fields like healthcare and medical research. At NYU Langone Health, we are actively exploring and integrating generative AI, ensuring its application remains a Compliant Clinical Tool For Health Care and research within our secure environment.
Upholding Legal Compliance with Generative AI in Healthcare Settings
In leveraging the power of generative AI, strict adherence to legal and ethical guidelines is non-negotiable. For healthcare professionals and researchers at NYU Langone Health, using AI responsibly means understanding and respecting critical boundaries. It is imperative to recognize that public generative AI applications are not compliant clinical tools for health care when it comes to sensitive data. Therefore, the following restrictions are in place to maintain patient privacy and data security:
- No Public AI for Clinical Documentation: Under no circumstances should public generative AI tools be used to create clinical documentation. This includes any notes intended for medical records or patient correspondence.
- Protection of Health Information (PHI): It is strictly forbidden to use public generative AI with Protected Health Information (PHI), even if anonymized, or any other legally protected data. This precaution ensures that patient confidentiality is maintained according to HIPAA and other relevant regulations.
- Clinical and Research Data Security: Similarly, public generative AI platforms must not be employed with clinical or human subjects research data, regardless of de-identification efforts. The integrity and privacy of research participants’ data are paramount.
- Confidential Business Information: Do not disclose any confidential business information to public generative AI applications. This protects NYU Langone Health’s proprietary information and strategic interests.
- Meeting and Activity Confidentiality: Public generative AI applications should not be permitted to record or upload recordings of internal meetings or any non-public activities at NYU Langone Health, safeguarding internal communications and strategic discussions.
- Verification of AI-Generated Content: It is crucial to understand that generative AI outputs from public platforms should not be directly relied upon without independent verification. NYU Langone Health professionals are responsible for validating any AI-generated content to ensure accuracy and appropriateness for clinical or research use. Public platforms are not designed as compliant clinical tools for health care and thus require careful scrutiny.
Accessing Secure and Compliant AI Tools at NYU Langone Health
Recognizing the transformative potential of AI while prioritizing security and compliance, NYU Langone Health provides access to a private generative AI environment. This controlled access ensures that professionals can utilize AI as a compliant clinical tool for health care within a secure and regulated framework.
- Rolling Access and Prioritization: Access to NYU Langone’s private generative AI environment is granted on a rolling basis. This approach helps manage system load and ensures a high-quality user experience for everyone.
- Prioritization Criteria: Access priority is given to individuals involved in innovation or research projects, mentored exploration projects, and general exploration projects. This prioritization aims to foster AI innovation in areas where it can significantly benefit healthcare and research.
- Data Use Agreement and Onboarding: Approved users receive email notifications and must sign a data use agreement. They are also required to review detailed information on how to access and effectively use our managed AI instance. This step ensures that all users are fully aware of their responsibilities and the secure protocols in place.
Guidelines for Utilizing Approved and Compliant AI Tools
To ensure the responsible and secure integration of AI in healthcare and research practices at NYU Langone Health, specific guidelines and usage policies are in place for our compliant clinical tool for health care:
- Approved AI Tool Platform: NYU Langone Health provides a designated and approved GPT platform. To gain access and utilize this compliant clinical tool for health care, applications are available through this link. This platform is designed to meet our stringent security and compliance requirements.
- Data Security Protocols: Before incorporating any sensitive data into AI programs, it is mandatory to obtain institutional permissions. This step ensures that data handling aligns with all regulatory and institutional policies, maintaining the highest standards of data protection.
- Avoidance of Unapproved Tools: To strictly maintain data security and ownership, the use of external chatbots or other non-approved AI tools for NYU Langone Health work is prohibited. Relying on our approved platform is crucial for using a truly compliant clinical tool for health care.
- Explore NYULH AI Resources: To learn more about how NYU Langone Health is strategically using generative AI and to understand best practices, further information is available at NYU Langone Health Generative AI Information. This resource provides valuable insights into navigating the AI landscape within our institution.
For any inquiries regarding AI initiatives at NYU Langone Health, please reach out to our dedicated team at [email protected]. We are committed to facilitating the responsible and innovative use of compliant clinical tools for health care to enhance patient care and advance medical research.
