Implementing AI Across the Enterprise
Artificial Intelligence (AI) has the potential to revolutionize the way we work. Johns Hopkins recognizes the immense potential of AI and offers a range of tools and services to harness its power. However, it is essential that these tools and services are used responsibly to ensure that the benefits of AI are realized for all.
Johns Hopkins-Approved AI Tools
Work Support
Clinical Tools Currently Under Exploration
More tools coming soon.
Guidelines for Using AI-Based Tools
Interdisciplinary teams of stakeholders from Johns Hopkins Medicine and Johns Hopkins University are meticulously curating guidelines for responsible integration of AI across the enterprise. With a deliberate and thoughtful approach, these groups are navigating the complexities of this evolving technological landscape.
As these groups work to curate more specific guidelines and recommendations, please review the general guidelines below before engaging with or promoting the use of a new AI tool.
Risk Management
- Remember AI results are only as good as their prompts, so human review of AI answers is requisite; AI can make mistakes!
- Manage risks associated with AI implementation, including legal, financial, and reputational risks
Ethics, Equity, & Accessibility
- Consider privacy, data security, bias mitigation, and transparency
- Respect and promote diversity and inclusion of different groups and consider the potential impact of AI on healthcare access, affordability, and equity
Data Governance
- Implement robust data governance policies to ensure the quality, integrity, and privacy of data used for development and deployment
- Maintain compliance with regulatory agencies and policies while enabling cross-disciplinary data sharing and collaboration
Privacy & Ethical Data Use
- Protect the personal data and information of individuals and organizations
- Respect the consent and preferences of data subjects
Transparency & Interpretability
- Prioritize development of AI models that provide reasoning behind AI-driven decisions
- Ensure models produce results that are easily understood
Validation & Regulation
- Validate AI algorithms rigorously through clinical trials and real-world testing to assess their safety, efficacy, and generalizability before widespread deployment
- Engage with regulatory bodies to ensure compliance with relevant regulations
Continuous Monitoring & Improvement
- Detect and address issues such as biases, errors, and performance degradation over time
- Incorporate feedback from faculty, staff, students, patients, and clinicians to iteratively improve AI algorithms
Interdisciplinary Collaboration
- Foster collaboration between researchers, clinicians, data scientists, and other relevant stakeholders
- Ensure AI solutions address the enterprise’s needs effectively and responsibly
Education & Training
- Offer education and training programs covering topics such as AI fundamentals, healthcare applications of AI, ethical considerations, and best practices for AI implementation
- Promote a culture of lifelong learning and interdisciplinary collaboration
Featured Case Studies
-
Case Study: Copilot for M365 in Excel
Copilot can help standardize and deduplicate data.
-
Case Study: Copilot for M365 in Outlook
Use Copilot in Outlook to summarize lengthy email threads.
-
Case Study: Copilot for M365 in Teams
Copilot can summarize Teams meetings and identify action items.
