HITRUST recently released a new assessment catering to AI security. Building on the HITRUST approach, it provides high-level assurance and certifies an organization’s commitment to robust, continuously improving cyber defenses in the face of evolving threats related to AI technology.
Is your organization ready for HITRUST certification? Schedule a consultation to find out.
Understanding AI Security with HITRUST
HITRUST’s new AI security module is the latest addition to its robust suite of cybersecurity assurance offerings. It gives organizations a robust, systematic way to assess and improve AI security practices as AI tools become more ingrained in the fabric of business operations and AI threats continue to evolve. These assessments are a means of future-proofing efficiently.
Appreciating the full value these assessments provide means understanding:
- How HITRUST approaches cybersecurity as a whole
- What HITRUST models are designed for AI security
- Which HITRUST AI framework is best in which cases
- What other assessments are available from HITRUST
Working with a trusted compliance advisor and HITRUST assessor is the best way to select the right HITRUST assessment for your organization, prepare for it, and conduct it successfully.
The HITRUST Approach to Cybersecurity
HITRUST is a leading cybersecurity standardization and certifying body. Born out of a focus on the healthcare industry, HITRUST has since broadened its scope and now provides guidance to organizations in every industry around the world. The approach HITRUST uses revolves around combining and harmonizing elements from several other frameworks and regulatory bodies into a single, omnibus framework that covers every conceivable need or use case for cyberdefense.
In particular, the HITRUST CSF is the primary model on which all of HITRUST’s assessments and services are based. The CSF is informed by dozens of cybersecurity frameworks. These include governmental regulations like the National Institute of Standards and Technology (NIST) Special Publication (SP) 800-53 and the General Data Protection Regulation (GDPR). They also include industry-specific rulesets like the Health Insurance Portability and Accountability Act (HIPAA) and the Payment Card Industry (PCI) Data Security Standards (DSS), along with global gold standards like the International Organization for Standardization (ISO)’s 27001 and 27002.
The CSF harmonizes thousands of requirements and specifications from across these sources into unified controls. The goal is for organizations to “assess once, report many”—avoiding costly overlap while maximizing security. This is why 75% of Fortune 20 companies use the HITRUST CSF, and over 99% of HITRUST-certified environments evaded breaches in 2024.
Looking ahead, organizations that apply HITRUST’s general approach to cybersecurity to their AI systems will be best-positioned to navigate the uncertainties of this emerging technology.
HITRUST Certifications Related to AI
Given the comprehensiveness of the CSF and its many sources, there have been elements of AI security embedded in it for years. Organizations that undergo a robust CSF implementation are already safeguarding themselves against many of the threats that AI brings. However, many but not all of these existing HITRUST protections addressed AI indirectly, building defenses that just so happened to apply to AI tools or mitigate AI threats rather than targeting them explicitly.
HITRUST has made a concerted effort in recent years to ramp up its AI-specific offerings, with targeted controls and vulnerability assessments that address the root causes inherent to AI.
At present, there are two primary models HITRUST offers for AI security assurance:
The risk management suite was developed first, and it remains the more basic of the two options. The security assessment and certification model, released in late 2024, is the most robust option presently available from HITRUST for securing AI systems organization-wide.
Let’s take a close look at each before comparing them for different business use cases.
HITRUST AI Risk Management Assessment
This is the entry-level AI security option offered by HITRUST. Envisioned as a turnkey solution, it enables organizations that are AI users to ensure that their risk management approach (i.e., risk tolerance and mitigation) is commensurate with the specific AI tools and systems they’re using, along with AI risk factors in their IT environments. The targeted implementation comprises up to 51 controls based on best practices from the CSF, ISO 23894, NIST AI Risk Management Framework, and others. This assessment is for organizations who use AI, but do not develop or provide AI to other organizations.
The security risk assessment is effective because all elements of security are informed by AI risks. These include both vulnerabilities within AI tools and systems themselves and new threat vectors and actors arising from the evolving ways that cybercriminals use AI to launch attacks.
In practice, organizations implement up to 51 controls, tailored to their AI dynamics. Then, they perform self-assessments using HITRUST MyCSF, an integrated software as a service (SaaS) platform that enables intricate reporting and analysis. Successful assessment results in an AI risk Management Insights Report detailing strengths, potential weaknesses, and elements to monitor in the short- and long-term. Organizations also receive one year of access to MyCSF.
The benefits of this report revolve around accessibility and efficiency; it creates easily sharable data especially useful for startups and smaller companies earlier on in their security journeys. It is critical to note here that the AI risk management assessment is not a certified or certifiable solution, it is simply a report.
HITRUST AI Security Assessment and Certification
The HITRUST AI Assurance Program represents a significantly more robust approach to AI security. It validates the overall security of AI systems by assessing risk management alongside other domains like infrastructure, access control, and incident response. The program provides a flexible yet rigorous framework for organizations that develop or deploy operational AI systems—rather than simply consume them—to demonstrate that their AI platforms are trustworthy, resilient, and secure.
Importantly, the AI Security Assessment is not intended for every organization using AI tools. The AI Security Assessment applies to those who develop, deploy, or operate AI systems as part of their infrastructure or offerings—not organizations that simply consume AI-enabled software or services. If your organization is responsible for integrating or managing operational AI models, the assessment is likely applicable. If you only use AI tools without altering or overseeing their underlying systems, the assessment is not required.
HITRUST awards AI security certification when organizations meet two conditions: (1) the curated set of AI-specific security requirements—tailored to the organization’s platform based on threat modeling and authoritative standards—reaches the required maturity thresholds, and (2) the underlying HITRUST assessment (e.g., e1, i1, or r2) is also certified. Organizations can add AI assessments to any HITRUST certification project. While most pair them with i1 or r2 assessments, some combine them with the lighter-weight e1, allowing flexibility based on cybersecurity maturity. This approach addresses AI-specific risks alongside foundational security practices.
The ai1 and ai2 certification include a curated set of AI-specific security requirements, tailored to your AI platform’s architecture and risk exposure. The number and nature of applicable controls are determined by system context and AI threat modeling, meaning no two assessments are exactly alike. This scope-based approach ensures that the controls align with the real-world risks relevant to your specific AI deployment.
As with other HITRUST evaluations, ai1 and ai2 certifications are conducted via the MyCSF platform. This streamlines the assessment process by enabling flexible, self-guided assessments and supporting third-party validation where needed. Automated evidence collection and integrated workflows improve audit readiness and reduce assessment fatigue.
Another key benefit is the ability to inherit AI controls from certified cloud service providers. If your AI systems are hosted on platforms that are already HITRUST certified, you can leverage their existing security controls. This helps reduce redundancy, simplifies integration, and mitigates third-party risks while maintaining high assurance levels.
Download Our HITRUST Checklist
Core HITRUST Certifications
AI security is critical, but it depends on broader, system-wide cyber defense. Protecting AI tools and defending against AI-related threats requires a comprehensive approach. To that effect, organizations pursuing an AI security assessment must also complete a core HITRUST certification—such as the e1, i1, or r2. This is because the AI assessment is designed to layer on top of these foundational certifications. As a result, these core HITRUST CSF assessments ensure that the broader IT environment meets essential cybersecurity standards. Thus, creating a secure baseline for evaluating and certifying AI-specific risks. This integrated model enables organizations to validate both general and AI-specific security in a single, streamlined process.
The mainline HITRUST certification assessments available at present are:
- HITRUST e1: Foundational Cybersecurity – This is a one-year validated assessment comprising 44 security controls. It’s an excellent entry point to broader security and sets startups and other smaller organizations up for other, more comprehensive deployments.
- HITRUST i1: Leading Security Practices – This one-year validated assessment builds on the e1 with the potential to span two years. It begins with 182 controls on average in Year 1, with an additional ~60 for rapid re-assessment in Year 2 (if implemented). The i1 is ideal for organizations that need robust, efficient security and compliance assurance.
- HITRUST r2: Expanded Practices – This two-year validated assessment provides the most robust and flexible cybersecurity assurance, with up to 250 controls and about 360 individual requirement statements on average and coverage for most compliance and regulatory needs. It’s ideal for large, mature organizations with complex risk exposure.
Beyond these, HITRUST also offers Insights Reports that facilitate mapping between CSF controls and several other frameworks, like HIPAA. There’s also a HITRUST NIST CSF 2.0 certification available as an add-on to r2 assessments for seamless government compliance.
Whatever HITRUST assessment is best for your organization, working with a compliance partner like RSI Security is the best way to prepare for, achieve, and maintain certification.
Rethink and Optimize Your AI Security Today
HITRUST has made significant strides in recent years to support both the secure use and responsible development of AI systems. The HITRUST AI Assurance Program includes two distinct assessment types, tailored to different stakeholders in the AI ecosystem.
Organizations that develop or deploy operational AI systems should consider the AI security assessment. Available in ai1 and ai2 formats, this assessment provides a certifiable framework to implement and validate system-specific controls. These assessments are layered on top of a certified HITRUST CSF assessment (e1, i1, or r2) to ensure that AI components are in alignment with broader cybersecurity infrastructure.
On the other hand, for organizations that use AI tools—but don’t build, deploy, or operate the systems themselves—HITRUST offers AI risk management assessments. Internal teams and third-party assessors use these non-certifiable tools to evaluate risks from AI integration. They deliver valuable insights into AI-related threats and guide organizations in prioritizing risk mitigation. AI risk management assessments don’t result in certification. However, they offer an ideal solution for startups and smaller organizations. These tools help teams conduct structured evaluations without pursuing full certification.
Together, the AI risk management and AI security assessments empower both AI users and providers. They help organizations align with authoritative security standards. They also support effective risk management and promote responsible AI adoption across diverse operational contexts.
RSI Security has helped countless organizations rethink and optimize their approach to AI and general information security. We’re trusted HITRUST partners who will facilitate assessments of any kind. We believe that discipline up-front unlocks freedom down the line, especially when implementing emerging technologies like AI. We’ll help you do so securely and efficiently.
To learn more about HITRUST AI assessments with RSI Security, get in touch today.
Contact Us Now!
