Navigating Illinois AIVIA: Your Guide to AI Impact Assessments and AI Compliance

The rapid integration of Artificial Intelligence (AI) into critical business functions, particularly human resources, presents both opportunities and significant regulatory challenges. For companies leveraging AI in their hiring processes, understanding and complying with state-specific regulations is paramount. The Illinois AI Video Interview Act (AIVIA), Bill HB 2557, effective January 1, 2020, stands as a foundational piece of legislation in this evolving landscape. While AIVIA doesn't explicitly mandate a formal "AI impact assessment" in the same vein as the upcoming Colorado AI Act or the EU AI Act, its stringent requirements for notice, consent, explanation, and data deletion effectively necessitate a comprehensive internal review process that mirrors the objectives of an impact assessment. This guide will walk compliance officers, general counsel, and CTOs through conducting a robust, AIVIA-focused impact assessment to ensure AI compliance and mitigate risks of algorithmic discrimination.

The stakes are high. Non-compliance with AIVIA can result in penalties of up to $1,000 per violation, enforced by the Illinois Department of Commerce and Economic Opportunity. As the regulatory environment for AI intensifies—with new laws like the Colorado AI Act (SB 24-205) coming into effect on June 30, 2026, and the EU AI Act (Regulation (EU) 2024/1689) imposing high-risk system obligations by August 2, 2026—proactive and thorough assessment is no longer optional but a strategic imperative. Leveraging an AI compliance platform like AICompliant can streamline this complex process, ensuring your organization not only meets AIVIA's demands but also builds a scalable AI risk management framework for the future.

Understanding the Illinois AIVIA: Your Foundation for AI Compliance

The Illinois AI Video Interview Act (AIVIA) applies to employers who "solely rely" on AI to analyze video interviews of applicants for an employment position. This specifically targets situations where AI is used to assess candidates' facial expressions, mannerisms, or other attributes without human review contributing to the initial screening decision.

AIVIA mandates four key requirements:

Notice: Employers must notify applicants that AI will be used to analyze their video interviews. This notice must inform applicants of the specific role of the AI in the hiring process.
Consent: Employers must obtain explicit consent from applicants to use AI for their video interview analysis. Applicants must be informed they can request that an alternative, non-AI analysis method be used if available, or decline the AI-driven interview without penalty (if an alternative is available).
Explanation: Upon request, employers must provide applicants with information about the characteristics the AI considers and how it informs their hiring decisions. This is where the implicit need for an "AI impact assessment" truly emerges, as employers must understand their AI system's inner workings to articulate this explanation clearly and accurately.
Deletion: Employers must delete all video interviews and associated data within 30 days of an applicant's request. This emphasizes data privacy and control, requiring robust data governance protocols.

While AIVIA doesn't explicitly use the term "AI impact assessment," adhering to its explanation requirement demands a deep understanding of the AI's functionality, its potential biases, and its impact on hiring outcomes. This necessitates a structured approach to evaluate the AI system—a process that is, in essence, an impact assessment focused on algorithmic discrimination prevention.

Compare this to other emerging regulations: NYC Local Law 144 of 2021, effective July 5, 2023, for Automated Employment Decision Tools (AEDTs), explicitly requires bias audits and public reporting. Maryland AI Employment Law (HB 1106), effective October 1, 2025, sets requirements for notifying employees about the use of AI in employment decisions. These laws, along with AIVIA, highlight a growing trend in state-level AI regulation, emphasizing the need for a comprehensive US state AI law tracker to manage diverse compliance requirements.

The De Facto AI Impact Assessment for AIVIA Compliance

To meet the spirit and letter of AIVIA, organizations must go beyond simply providing notice and consent. They must understand how their AI system makes decisions, what data it relies on, and whether it introduces or exacerbates biases. This internal review process, though not formally named an "impact assessment" by AIVIA, serves the same critical function: identifying, assessing, and mitigating risks associated with the AI system.

This process draws heavily on principles found in broader AI governance frameworks, such as the NIST AI Risk Management Framework (AI RMF 1.0) and ISO/IEC 42001:2023. These frameworks provide a structured approach to identifying AI risks, including fairness, privacy, and accountability, and are often referenced by regulators as best practices. By applying a similar structured approach, even for AIVIA, companies can establish a consistent AI risk management framework that scales to more prescriptive future regulations.

The core objective of this assessment is to ensure the AI-powered video interview tool is fair, transparent, and accountable. It's about proactively addressing the potential for bias and ensuring that the explanations provided to candidates are genuine and verifiable.

Step-by-Step Guide to Conducting Your AIVIA-Focused AI Impact Assessment

This step-by-step guide outlines the critical components of an AIVIA-focused AI impact assessment, enabling your organization to establish defensible AI compliance requirements by state while setting the stage for broader regulatory adherence.

Step 1: Define the AI System and Its Purpose

Begin by precisely identifying the AI system used for video interview analysis. This includes:

System Identification: Name of the AI tool/platform (e.g., vendor solution, in-house developed).
Intended Purpose: Clearly articulate why the AI is being used in hiring (e.g., initial screening, sentiment analysis, skill assessment). What specific job roles or stages of the hiring process does it impact?
Key Features and Functionality: Detail what the AI purports to do. Does it analyze speech patterns, facial expressions, keyword usage, or a combination? How does it generate its "scores" or recommendations?
Decision-Making Impact: Document how the AI's output is used in the overall hiring decision-making process. Is it a recommendation, a filter, or a primary driver? Does a human review always occur before a final decision?
Accountability: Identify the internal stakeholders responsible for the AI system's deployment, monitoring, and compliance.

Step 2: Identify and Document Data Sources

Understanding the data feeding your AI is crucial for identifying potential biases and privacy concerns.

Training Data: What data was used to train the AI model? Origin, volume, demographics, and characteristics of this data are vital. Inadequate or biased training data is a primary source of algorithmic discrimination. Be mindful of regulations like California AB 2013 (Training Data), effective January 1, 2025, which mandates transparency regarding training data for specific AI models. Penalties for non-compliance can reach up to $7,500 per violation.
Input Data: What data does the AI system collect from applicants during the video interview (e.g., video, audio, transcription, metadata)?
Data Collection Methods: How is this data acquired and stored?
Privacy Implications: Assess whether the data collection aligns with your company's privacy policy and other data protection regulations (e.g., GDPR, if applicable, with penalties up to $20,000,000 per violation).
Data Governance: Document data retention policies, access controls, and deletion procedures, especially in light of AIVIA's deletion requirement.

Step 3: Assess AI Capabilities, Limitations, and Potential Biases

This is the core of your AIVIA-focused impact assessment and crucial for algorithmic discrimination prevention. It requires a deep dive into the AI's analytical mechanisms.

Algorithm Transparency: Work with your vendor or internal AI team to understand the "black box." What specific features or characteristics does the AI analyze? How are these weighted? This information is essential for fulfilling AIVIA's explanation requirement.
Bias Identification: Conduct thorough bias testing. This involves:
- Disparate Impact Analysis: Does the AI system's output (e.g., hiring recommendations) disproportionately favor or disfavor certain demographic groups (race, gender, age, disability status)?
- Fairness Metrics: Utilize recognized fairness metrics (e.g., demographic parity, equal opportunity, predictive equality) to evaluate the AI's performance across different groups.
- Data Shift Analysis: Test if the AI's performance degrades or shifts unfairly when presented with different accents, lighting conditions, or cultural expressions.
Limitations: Document the AI's known limitations or areas where its performance is suboptimal or unverified. For example, is it less accurate for non-native English speakers or individuals with certain speech impediments?
Practical Tips:
- Consult Vendors: Request comprehensive documentation from your AI vendor regarding their bias testing methodologies, fairness reports, and transparency statements.
- Internal Testing: Conduct your own internal audits and shadow testing, comparing AI recommendations with human decisions across diverse applicant pools.
- Third-Party Audits: Consider engaging independent experts to perform a bias audit, similar to requirements under NYC Local Law 144.
Leveraging AICompliant: An AI compliance platform like AICompliant can centralize bias testing results, track mitigation efforts, and provide a structured repository for vendor documentation, significantly easing the burden of this complex step. The platform's /tools/compliance-checker can assist in systematically reviewing various AI risks, including bias.

Step 4: Evaluate Risks and Harms

Based on the assessment in Step 3, identify potential harms that could arise from the AI's use, especially in the context of AIVIA.

Discrimination: Direct or indirect discrimination based on protected characteristics (e.g., race, gender, age, disability, national origin) is the primary risk.
Loss of Opportunity: Applicants unfairly screened out due to AI bias.
Privacy Violations: Improper handling or storage of video and biometric data.
Reputational Damage: Public backlash, negative press, and loss of trust if AI is found to be discriminatory.
Legal & Financial Risks: Penalties for non-compliance (AIVIA: up to $1,000 per violation), legal fees, and potential class-action lawsuits.

Step 5: Implement Mitigation Strategies

Once risks are identified, develop and implement concrete strategies to mitigate them.

Bias Mitigation Techniques:
- Data Pre-processing: Address biases in training data before model training.
- Algorithm Adjustments: Implement fairness-aware algorithms.
- Post-processing: Adjust AI outputs to reduce disparate impact.
Transparency Mechanisms:
- Clear Disclosure: Ensure AIVIA's notice requirement is met with unambiguous language.
- Detailed Explanation: Prepare a standardized, yet comprehensive, explanation of how the AI functions for applicants who request it. This can be greatly facilitated by maintaining an up-to-date AI impact assessment template.
Human Oversight: Ensure human review and override capabilities are always in place. The AI should augment human decision-making, not replace it entirely, particularly in high-stakes contexts like hiring.
Data Retention & Deletion: Establish robust, automated processes to handle applicant data deletion requests promptly and securely, adhering to AIVIA's 30-day requirement.
Feedback Loops: Implement mechanisms for applicants to provide feedback on the AI process.

Step 6: Documentation and Ongoing Monitoring

Thorough documentation is the backbone of defensible AI compliance. It demonstrates due diligence and provides evidence for regulators, should it ever be required.

Comprehensive Records: Maintain detailed records of your AI impact assessment, including methodologies, findings, mitigation strategies, and outcomes. This documentation forms a vital part of your AI risk management framework.
Policy & Procedure Updates: Integrate AIVIA compliance requirements into your existing HR policies, data privacy policies, and vendor management procedures.
Regular Audits: Conduct periodic internal and external audits of your AI systems to ensure continued fairness, accuracy, and compliance. An AI compliance checklist 2026 should be regularly updated to reflect evolving regulations.
Continuous Monitoring: Establish a system for ongoing monitoring of AI performance and fairness metrics. This is especially crucial as AI models can drift over time.
Reassessment Triggers: Re-evaluate your AI system's impact assessment if there are significant changes to the AI model, its intended use, the data it processes, or relevant regulations.
AICompliant's Role: An AI compliance software like AICompliant excels at centralizing this documentation, automating audit trails, and providing a dynamic /dashboard for continuous monitoring. It transforms static assessments into an active, manageable compliance program, ensuring your AI compliance requirements by state are met consistently.

Leveraging AICompliant for AIVIA and Broader AI Compliance

Managing the intricacies of AIVIA compliance, alongside the rapidly expanding web of other state and international AI regulations, can overwhelm even the most sophisticated legal and compliance teams. This is where an advanced AI compliance platform becomes indispensable.

AICompliant offers an end-to-end solution designed to automate and simplify AI governance. It provides a structured environment to:

Automate AI Impact Assessments: Leverage pre-built templates and guided workflows to conduct your AIVIA-focused assessments, ensuring all critical steps are covered and documented. Our /tools/compliance-checker can kickstart your assessment process.
Centralize Documentation: Keep all AI system details, data sources, bias test results, mitigation strategies, and policy documents in a single, secure repository.
Track Regulatory Changes: Stay ahead of the curve with an integrated US state AI law tracker that alerts you to new requirements, such as the Colorado AI Act's effective date of June 30, 2026, or the Texas Responsible AI Governance Act (TRAIGA), Bill HB 149, effective January 1, 2026, with penalties up to $200,000 per violation.
Monitor and Report: Utilize the /dashboard for real-time insights into your AI systems' compliance status, facilitating continuous monitoring and audit readiness.
Ensure Algorithmic Discrimination Prevention: AICompliant helps embed fairness and transparency principles throughout your AI lifecycle, actively working towards preventing biased outcomes in hiring.

By utilizing AICompliant, companies can transform a complex, manual process into an efficient, automated AI compliance workflow, allowing teams to focus on strategic oversight rather than administrative burden.

The Future of AI Regulation: Beyond AIVIA

While AIVIA provides a clear framework for AI in video interviews, it's just one piece of a much larger and rapidly evolving regulatory puzzle. Organizations must adopt a holistic approach to AI risk management framework development to remain compliant.

Consider the following upcoming and existing regulations:

Colorado AI Act (SB 24-205): Effective June 30, 2026, this landmark legislation imposes duties of care and transparency requirements on developers and deployers of high-risk AI systems, carrying penalties of up to $20,000 per violation. It also explicitly mandates impact assessments for high-risk AI.
EU AI Act (Regulation (EU) 2024/1689): Effective August 1, 2024, with high-risk system obligations kicking in by August 2, 2026, this is one of the world's most comprehensive AI laws. It mandates extensive conformity assessments, risk management systems, human oversight, and robust documentation for high-risk AI systems (which often include employment AI). Penalties can be as high as $35,000,000 per violation.
California AI Transparency Act (SB 942): Effective January 1, 2026, with penalties up to $5,000 per violation per day, requiring clear disclosures for AI-generated content.
Maryland AI Employment Law (HB 1106): Effective October 1, 2025, with penalties up to $10,000 per violation, requiring employers to notify applicants and employees about the use of AI in hiring and promotion decisions.

Developing an AI risk management framework that is adaptable and forward-looking, supported by robust AI compliance software, is essential for navigating this complex future. Your AIVIA-focused assessment is not just about Illinois; it's a foundational step towards broader AI compliance checklist 2026 readiness.

Conclusion

The Illinois AI Video Interview Act (AIVIA) represents an early but significant legislative effort to regulate AI in hiring. For organizations leveraging AI-powered video interviews, a structured internal review—akin to an AI impact assessment—is not merely good practice but a necessary component of robust AI compliance. By meticulously defining your AI system, understanding its data, assessing biases, implementing mitigation strategies, and maintaining rigorous documentation, you can meet AIVIA's requirements, avoid penalties, and foster a fairer hiring environment.

Proactive compliance, however, extends beyond current mandates. As the regulatory landscape continues to evolve, embracing an automated AI compliance solution like AICompliant will be crucial. It empowers your organization to manage complex requirements, mitigate risks effectively, and build a sustainable AI risk management framework that stands the test of time and innovation.

Take the Next Step Towards AI Compliance Excellence

Ready to streamline your AI compliance efforts and ensure your organization adheres to AIVIA and the growing global AI regulatory framework? Discover how AICompliant’s powerful platform can help you conduct comprehensive AI impact assessments, manage risks, and automate your compliance workflows.

Visit https://aicompliant.ai/pricing today to learn more and secure your customized solution.

Frequently Asked Questions

What is the Illinois AI Video Interview Act (AIVIA)?

The Illinois AI Video Interview Act (AIVIA), Bill HB 2557, effective January 1, 2020, is a law that regulates employers' use of AI to analyze video interviews of job applicants. It requires employers to provide notice, obtain consent, offer explanations about how the AI works, and allow for the deletion of video interview data.

Does AIVIA require a formal AI Impact Assessment like the EU AI Act?

AIVIA does not explicitly use the term "AI impact assessment." However, to comply with its requirements, particularly the need to explain how the AI works and what characteristics it considers, employers must conduct a detailed internal review process that functions as a de facto AI impact assessment. This involves understanding the AI's capabilities, limitations, and potential for algorithmic discrimination.

What are the penalties for non-compliance with the Illinois AIVIA?

Non-compliance with the Illinois AI Video Interview Act can lead to penalties of up to $1,000 per violation. Enforcement is handled by the Illinois Department of Commerce and Economic Opportunity.

How does AICompliant help with AIVIA compliance?

AICompliant serves as an AI compliance platform that streamlines the entire process. It offers tools for conducting structured AI impact assessments, centralizing documentation, automating compliance workflows, tracking regulatory changes (including those for AIVIA and other state laws), and providing a dashboard for continuous monitoring, helping organizations achieve automated AI compliance.

What other significant US state AI laws should companies be aware of besides AIVIA?

Beyond AIVIA, companies should monitor the Colorado AI Act (SB 24-205), effective June 30, 2026, which mandates impact assessments for high-risk AI; NYC Local Law 144, effective July 5, 2023, for Automated Employment Decision Tools; and the Maryland AI Employment Law (HB 1106), effective October 1, 2025. These laws, along with others like the California AI Transparency Act (SB 942) effective January 1, 2026, underscore the importance of a robust US state AI law tracker and a comprehensive AI risk management framework.

Illinois AIVIA: AI Impact Assessment & AI Compliance