Responsible AI

AI Policy

Effective Date: June 25, 2026

IurisIQ uses artificial intelligence to help attorneys work faster and more accurately. This policy explains exactly how AI works inside our platform, what we do and don't do with your data, and what you need to know as a licensed professional using AI-assisted tools.

We wrote this in plain language because we think you deserve a straight answer — not a legal hedge.

The short version: Your client data is yours. We never use it to train AI models. Our AI assists attorneys — it does not replace attorney judgment. Every AI output requires your review before it goes anywhere.

1. What AI Does in IurisIQ

AI features in IurisIQ are designed to reduce repetitive work and surface information faster. Current AI-assisted capabilities include:

Document Drafting Assistance

The AI drafting tool generates first-draft letters, motions, immigration cover letters, and client communications based on case data you have already entered in the platform. It does not retrieve or generate information from outside your case file.

Conflict of Interest Check

AI assists in identifying potential conflicts by scanning matter and party data across your case database. The system surfaces likely matches and flags them for attorney review — it does not make conflict determinations. That judgment remains entirely yours.

Document Translation

Built-in AI translation converts documents, intake forms, and correspondence between English and Spanish (with additional languages available on request). Translations are generated for reference and attorney review — they are not certified translations.

Case Intelligence

The AI Brain module analyzes case data to surface deadlines, flag missing documents, identify status patterns, and suggest next steps. Recommendations are informational — the attorney decides what action to take.

Intake Form Analysis

AI can pre-fill portions of matter records from completed intake forms, reducing manual data entry. All pre-filled data is presented for attorney or staff review and confirmation before being committed to the file.

2. What We Do — and Don't Do — With Your Data

This is the question attorneys ask us most. Here is the complete picture:

WE DO
  • ✓ Process your data to deliver AI features within your account
  • ✓ Keep your data isolated in your firm's private tenant
  • ✓ Use aggregated, anonymized platform telemetry to improve performance
  • ✓ Allow you to opt in to AI feature improvements with explicit consent
WE DO NOT
  • ✗ Use your client data to train AI models — ever
  • ✗ Share your case data with third-party AI providers for training
  • ✗ Allow AI outputs from one firm's data to influence another firm's results
  • ✗ Sell or license your data to any third party
  • ✗ Use AI to make autonomous decisions about your cases or clients

3. Private AI Architecture

IurisIQ operates on a single-tenant architecture. Your firm's data lives in its own isolated environment — it is never pooled with or accessible to other firms, even within shared infrastructure. This means:

  • AI features run against your data only — no cross-firm data leakage is architecturally possible
  • When AI models process your documents, that processing occurs within your tenant boundary
  • Any AI model we deploy is treated as a processor operating under strict data handling constraints, not as a general-purpose service with access to your content

Where IurisIQ uses third-party AI infrastructure (such as foundation model APIs for document generation), those providers are contracted under data processing agreements that prohibit training on customer content and require equivalent data isolation standards.

4. Attorney Responsibility and Human Oversight

AI-assisted features in IurisIQ are tools, not decision-makers. The following principles govern how AI interacts with attorney work product:

You must review all AI output before use

Every AI-generated draft, translation, conflict flag, and recommendation is presented to you for review before it is filed, sent, or acted upon. The platform is designed so that no AI output is automatically transmitted to a client, court, or government agency.

AI does not practice law

Nothing generated by IurisIQ's AI constitutes legal advice, a legal opinion, or a legal determination. As a licensed attorney, you remain solely responsible for the accuracy, completeness, and professional appropriateness of all work product — including work product that was drafted or assisted by AI.

Competence and supervision obligations remain yours

Your state bar's rules of professional conduct apply to your use of AI tools. Many bars have issued formal guidance on attorney competence in AI use, including obligations to understand how AI tools work, supervise AI-generated output, and maintain client confidentiality. IurisIQ does not relieve you of any of those obligations. We recommend consulting your state bar's AI guidance before using AI-assisted features for client matters.

Verify AI output for accuracy

AI systems can produce outputs that appear plausible but are factually incorrect, out of date, or inappropriate for a specific matter — a risk sometimes called "hallucination." You must independently verify all AI-generated content for factual accuracy, legal correctness, and suitability before relying on it in any client matter, filing, or communication.

5. Immigration Filings and Government Submissions

When IurisIQ assists in preparing immigration filings or transmitting data through government APIs (including the USCIS API), the following additional rules apply:

  • You must review the complete contents of any filing before submission
  • AI-assisted form population is a drafting aid — the attorney of record is responsible for verifying that all information is accurate, complete, and consistent with the client's actual circumstances
  • Submitting inaccurate information to a government agency may constitute a violation of federal law and professional conduct rules regardless of whether AI was used in preparation
  • IurisIQ does not verify, certify, or take responsibility for the accuracy of any government filing prepared using the platform

6. Bias, Fairness, and Limitations

AI systems reflect patterns in the data they were trained on and can produce outputs that are biased, incomplete, or incorrect. IurisIQ takes the following steps to reduce this risk:

  • We evaluate AI features for disparate output quality across common client demographics before deployment
  • We do not use AI to make eligibility determinations, risk scores, or recommendations that affect whether a client should be represented
  • We maintain feedback channels so attorneys can report AI outputs that appear biased, incorrect, or harmful
  • We disclose known limitations of specific AI features within the platform at the point of use

Despite these measures, no AI system is free of bias or error. Attorney review of AI output is the final safeguard, and your professional judgment supersedes any AI recommendation.

7. Transparency and Labeling

IurisIQ labels AI-generated content within the platform so you always know what was produced by AI and what was entered by a human. We do not present AI output as human-authored without disclosure.

If you share AI-generated content with clients or third parties outside the platform, you are responsible for any disclosure obligations that may apply under applicable professional conduct rules or law.

8. Bar Association Guidance

A growing number of state bars and the American Bar Association have issued formal guidance on attorneys' ethical obligations when using AI tools. Key themes from that guidance include:

  • Competence (ABA Model Rule 1.1): Attorneys must understand the benefits and risks of technology they use, including AI
  • Supervision (ABA Model Rule 5.3): Attorneys must supervise AI tools as they would non-lawyer assistants
  • Confidentiality (ABA Model Rule 1.6): Attorneys must take reasonable steps to prevent unauthorized disclosure of client information through AI tools
  • Communication (ABA Model Rule 1.4): Some guidance suggests disclosure to clients when AI is used in their matter may be appropriate

IurisIQ is designed to support compliance with these obligations — single-tenant isolation addresses confidentiality; labeling and review gates support supervision; and this policy supports informed use. We recommend you consult your state bar's current AI guidance, as it is evolving rapidly.

9. Opting Out of AI Features

AI-assisted features in IurisIQ are optional. Firm administrators can disable specific AI features for their account at any time through the platform settings. Disabling AI features does not affect access to any other part of the platform.

If you have concerns about a specific AI feature or want to understand how a particular output was generated, contact our support team at [email protected].

10. Reporting AI Errors and Concerns

We take reports of AI errors seriously. If an AI feature produces output that is incorrect, biased, harmful, or otherwise problematic, please report it:

Include a description of what the AI produced, the feature used, and why you believe the output was problematic. We review all reports and use them to improve our AI systems.

11. Changes to This Policy

AI capabilities in IurisIQ will evolve. When we add new AI features or change how existing features work in a material way, we will update this policy and notify subscribers by email at least 30 days before changes take effect. The current version of this policy is always available at iurisiq.com/ai-policy.html.

12. Contact