In 2024, the U.S. Department of Health and Human Services (HHS) finalized a new federal regulation known as the HTI-1 Final Rule (Health Data, Technology, and Interoperability), which introduces transparency requirements for artificial intelligence and predictive algorithms used in certified health IT.
This landmark rule represents the first time that the Office of the National Coordinator for Health Information Technology (ONC) has regulated the use of AI and machine learning models in certified EHR systems. Under the HTI-1 rule, developers must disclose how predictive algorithms work, what data they use, how they are evaluated, and how they are governed—especially when those tools influence clinical decision-making.
In this blog post, we explain the core requirements of the HTI-1 Transparency Rule and show how the Pacific AI Governance Policy Suite enables organizations to comply with each provision. For each HTI-1 criterion, we map the corresponding Pacific AI policy and clause.
What Is the HTI-1 Transparency Rule?
The HTI-1 rule, issued under ONC’s certification program, applies to any certified health IT that includes predictive algorithms. These systems are widely used in clinical decision support, population health management, risk scoring, and more.
Key transparency requirements under HTI-1 include:
- Disclosure of the algorithm’s intended use, target population, and performance
- Information about training data, limitations, and evaluation methods
- Clear governance and monitoring processes
- Support for clinical review and human oversight
- Regular updates and documentation of post-deployment monitoring
The goal is to give clinicians and patients better insight into how AI systems operate—so that these tools can be used safely and responsibly in care settings.
How the Pacific AI Governance Policy Suite Maps to HTI-1 Requirements
The Pacific AI Governance Policy Suite is a modular set of operational policies designed to support compliance with U.S. laws and regulations governing AI. It includes specific clauses that address transparency, lifecycle management, risk, fairness, and safety.
Below is a mapping of HTI-1 transparency requirements to the relevant Pacific AI policy documents and clauses.
| HTI-1 Requirement | Pacific AI Policy | Clause |
|---|---|---|
| Disclosure of model limitations and risk of bias | AI Fairness Policy; AI Risk Management Policy | §6.1; §5.2 |
| Documentation of training data and population coverage | AI System Lifecycle Policy | §7.1, §7.3 |
| Explanation of evaluation metrics and monitoring results | AI Risk Management Policy | §5.1-5.3 |
| Human oversight and decision support role | AI Safety Policy | §4.1, §7 |
| Governance processes for model updates and versioning | AI System Lifecycle Policy | §6; AI Risk Management Policy – §6.3 |
| Communication of changes to users | AI Transparency Policy | §6.2 |
Detailed Example: Disclosing Predictive Tool Information in EHR Systems
Suppose a hospital implements a predictive model that flags patients at risk for sepsis in its certified EHR system. Under the HTI-1 rule, the hospital must be able to provide users with clear, structured documentation about how that model works.
The Pacific AI suite supports this need by requiring:
- A plain-language summary of model function and intended use (AI Transparency Policy §4.1)
- A documented record of training data, test data, and population limitations (AI System Lifecycle Policy §7.1)
- Ongoing monitoring reports with outcome tracking and risk analysis (AI Risk Management Policy §5.3)
- Clear instructions for clinician override and review (AI Safety Policy §4.1, §7)
These requirements help ensure that predictive algorithms in certified health IT are not black boxes—but transparent, auditable tools that clinicians can trust.
Conclusion
The HHS HTI-1 rule marks a major step forward in AI transparency in healthcare. For developers of certified health IT and providers who implement AI tools, the message is clear: if your system uses predictive algorithms, you need a policy framework that ensures transparency, oversight, and continuous evaluation.
The Pacific AI Governance Policy Suite provides exactly that. With policy clauses that align directly with HTI-1’s transparency criteria, the suite helps organizations:
- Create disclosures clinicians can understand
- Manage model updates safely and systematically
- Document oversight and mitigate risk
By adopting the Pacific AI suite, healthcare organizations can accelerate compliance, reduce the risk of enforcement, and improve trust in the AI systems they deploy.
Download the Pacific AI Policy Suite at https://pacific.ai
Need help aligning with the HTI-1 rule? Contact [email protected]