File size: 8,413 Bytes
8be56c1 893f951 8be56c1 893f951 f85eb47 8be56c1 893f951 8be56c1 f85eb47 8be56c1 f85eb47 8be56c1 893f951 f85eb47 893f951 f85eb47 893f951 f85eb47 8be56c1 f85eb47 893f951 f85eb47 8be56c1 f85eb47 8be56c1 f85eb47 8be56c1 f85eb47 8be56c1 f85eb47 8be56c1 f85eb47 8be56c1 f85eb47 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 | ---
license: mit
language:
- en
tags:
- hipaa
- healthcare
- compliance
- legal
- rag
- regulations
- mojar-ai
pretty_name: HIPAA Knowledge Base (Mojar AI)
size_categories:
- n<1K
viewer: false
---
# HIPAA Knowledge Base — Mojar AI
The curated knowledge corpus that powers the **[Mojar AI HIPAA Compliance Assistant](https://www.mojar.ai/free-tools/hipaa-compliance-assistant)** — a free, public AI agent built by [Mojar AI](https://mojar.ai) to help healthcare teams, compliance officers, and engineers get accurate, grounded answers to HIPAA questions.
This dataset is the _source of truth_ the assistant retrieves from. Every answer the agent gives is grounded in one of the documents below, with citations back to the underlying CFR section, HHS guidance page, or curated explainer.
## Why this exists
HIPAA is large, fragmented, and scattered across:
- 45 CFR Parts 160, 162, 164 (the regulations themselves)
- HHS Office for Civil Rights (OCR) guidance pages and FAQs
- OCR resolution agreements and enforcement actions
- HITECH Act amendments
- Real-world operational practice that _isn't_ written down anywhere official
A general-purpose LLM hallucinates citations, mixes up the Privacy Rule with the Security Rule, and gives advice that _sounds_ right but cites sections that don't exist. The Mojar AI HIPAA agent solves that by retrieving from this curated, version-controlled corpus and answering only from what it finds.
We're publishing the corpus openly so:
1. The HIPAA agent's answers are **auditable** — anyone can inspect the exact source material.
2. Other teams can build their own HIPAA-aware tools on the same foundation.
3. Researchers studying healthcare compliance, RAG grounding, or regulatory NLP have a clean, citation-faithful starting point.
## Try the assistant
→ **[mojar.ai/free-tools/hipaa-compliance-assistant](https://www.mojar.ai/free-tools/hipaa-compliance-assistant)**
Ask it questions like:
- "Can a covered entity use a free tier of a cloud provider for PHI?"
- "What's the difference between consent and authorization under HIPAA?"
- "When does the Breach Notification Rule require notifying the media?"
- "What technical safeguards does §164.312 require for access control?"
The agent retrieves from this dataset, cites the source, and refuses to answer when the corpus doesn't cover it.
## Contents
### `datasets/` — Regulatory & reference material
Long-form Markdown documents covering the HIPAA landscape:
| File | Topic |
| ------------------------------------------------ | ----------------------------------------------- |
| `01-hipaa-privacy-rule.md` | Privacy Rule (45 CFR 164 Subpart E) |
| `02-hipaa-security-rule.md` | Security Rule (45 CFR 164 Subpart C) |
| `03-breach-notification-rule.md` | Breach Notification Rule (45 CFR 164 Subpart D) |
| `04-phi-de-identification.md` | Safe Harbor & Expert Determination methods |
| `05-business-associates.md` | BAAs, subcontractors, liability |
| `06-phi-definitions-identifiers.md` | What counts as PHI; the 18 identifiers |
| `07-patient-rights.md` | Access, amendment, accounting of disclosures |
| `08-technical-safeguards-engineering.md` | §164.312 from an engineering POV |
| `09-real-world-compliance-pitfalls.md` | Common mistakes in practice |
| `10-cloud-computing-hipaa.md` | OCR cloud guidance, BAA gotchas |
| `11-consent-vs-authorization.md` | When each is required |
| `12-digital-marketing-tracking-risks.md` | Pixels, analytics, OCR tracking bulletins |
| `13-hitech-act.md` | HITECH amendments, breach floor |
| `14-enforcement-penalties.md` | CMP tiers, willful neglect |
| `15-hhs-faq-categories.md` | Index of HHS FAQs by topic |
| `16-rag-system-architecture.md` | Notes on building HIPAA-aware RAG |
| `17-covered-entities.md` | Who's regulated and who isn't |
| `18-minimum-necessary-standard.md` | §164.502(b) in practice |
| `19-45-cfr-part-160-verbatim.md` | Verbatim regulatory text |
| `20-45-cfr-part-162-verbatim.md` | Verbatim regulatory text |
| `21-45-cfr-part-164-verbatim.md` | Verbatim regulatory text |
| `22-hhs-official-guidance-pages-verbatim.md` | Verbatim HHS/OCR guidance |
| `23-hhs-faq-page-verbatim.md` | Verbatim HHS FAQ content |
| `24-ocr-resolution-agreements-index-verbatim.md` | Index of OCR enforcement actions |
| `24-q-and-a.md` | Compiled Q&A reference |
### `datasets-real-life/` — Role-based practitioner Q&A
Practical HIPAA Q&A grouped by who's asking:
- `01-doctors-qa.md` — clinicians at the point of care
- `02-nurses-qa.md` — bedside / floor staff
- `03-admin-frontdesk-qa.md` — reception, scheduling, intake
- `04-billing-insurance-qa.md` — RCM, payers, claims
- `05-compliance-privacy-qa.md` — privacy officers, security officers
- `200-role-based-hipaa-qa-summary.md` — consolidated 200-question reference
- `HIPAA_200_QA_Guide.md` — companion authoring guide
- `Building HIPAA Question Datasets.md` — methodology notes
### `tmp/`
Raw source material (e.g. `title-45-2026-04-24.xml` — the U.S. Code of Federal Regulations Title 45 export) used to build the curated documents above. Kept in the repo for reproducibility.
### `Building a HIPAA AI Agent.md`
Design notes describing how Mojar AI assembled the HIPAA agent on top of this corpus — chunking strategy, retrieval setup, citation enforcement, refusal behavior, and evaluation.
## Format
All documents are plain Markdown (`.md`), suitable for direct ingestion into RAG pipelines.
- **Verbatim files** (CFR, HHS guidance, FAQs, OCR resolution agreements) preserve original wording for citation fidelity. Do not paraphrase these into your index.
- **Curated explainers** are written in plain English for retrieval-friendly grounding. Each one cites the underlying CFR/HHS source at the top.
- **Role-based Q&A** files use a consistent `Q:` / `A:` format suitable for fine-tuning, evaluation sets, or example-based retrieval.
## Updates
The corpus is updated as HHS/OCR publishes new guidance and as we add operational Q&A from real customer usage of the assistant. Material changes are committed with a descriptive message; see the commit history.
## License
Released under the **MIT License**.
- U.S. federal regulations (45 CFR) and HHS/OCR guidance are public-domain works of the U.S. Government.
- Curated commentary, explainers, role-based Q&A, and architectural notes are © Mojar AI, licensed MIT.
## Disclaimer
This dataset and the [Mojar AI HIPAA Compliance Assistant](https://www.mojar.ai/free-tools/hipaa-compliance-assistant) are provided **for informational and educational purposes only**. They do **not** constitute legal advice, do not create an attorney–client relationship, and are not a substitute for guidance from qualified counsel or your organization's designated Privacy Officer / Security Officer. HIPAA compliance decisions should be made in consultation with appropriate professionals who can evaluate your specific facts and circumstances.
## About Mojar AI
[Mojar AI](https://mojar.ai) builds enterprise RAG and AI agents grounded in your organization's own documents — the same architecture that powers this public HIPAA assistant, applied to legal contracts, RFPs, internal policies, clinical guidelines, and any other domain where citation accuracy matters more than fluency.
- Website: [mojar.ai](https://mojar.ai)
- HIPAA assistant: [mojar.ai/free-tools/hipaa-compliance-assistant](https://www.mojar.ai/free-tools/hipaa-compliance-assistant)
|