Version 1.0 · July 10, 2026

1. Policy Purpose

This AI Policy establishes how Ravi Singh uses artificial intelligence responsibly, transparently, and safely in its professional influence products.

The purpose of this policy is to ensure that AI is used to provide structured professional feedback, improve the quality of profile and content analysis, generate personalized recommendations, support user self-reflection and professional development, protect users from misleading claims, reduce bias and hallucination risk, and maintain human accountability over product design, methodology, and customer-facing claims.

Ravi Singh does not use AI to make employment, credit, housing, education, lending, insurance, legal, medical, immigration, or government-benefit decisions.

2. Product Scope

This policy applies to all AI-supported features in the Ravi Singh professional influence ecosystem.

3. Core AI Principles

3.1 Transparency

Users must be informed when AI is used to generate or assist with reports, summaries, recommendations, rewrites, classifications, or scores.

3.2 Human Accountability

AI may assist with analysis, but Ravi Singh remains responsible for product design, scoring logic, prompt design, approved disclaimers, customer-facing claims, escalation procedures, quality assurance, refund or re-audit decisions, and methodology updates. AI output should never be treated as automatically correct.

3.3 No Misleading Claims

Ravi Singh must not claim that AI can guarantee employment, recruiter interest, promotion, follower growth, virality, income, investment outcomes, partnership opportunities, verified expertise, professional competence, legal compliance, or platform algorithm success.

3.4 Privacy and Data Minimization

The product should collect only the data needed to generate the requested assessment, audit, report, badge, credential, or verification output.

3.5 User Control

• Understand what data is being used.
• Correct inaccurate submitted information.
• Request deletion of stored report data where legally and operationally feasible.
• Know whether their report was generated by AI or fallback logic.
• Choose whether to share their badge, credential, or report publicly.

3.6 Accuracy and Reliability

The system should use structured scoring, validation, and version control to reduce inconsistency. AI-generated score and grade outputs should be versioned, validated, and clearly distinguished from any self-assessment score because they measure different things.

3.7 Fairness and Non-Discrimination

AI reports must not make assumptions or recommendations based on protected characteristics, including race, color, religion, national origin, sex, sexual orientation, gender identity, age, disability, veteran status, marital status, pregnancy, or genetic information.

4. Approved AI Use Cases

• Interpreting assessment scores.
• Generating report narratives.
• Explaining Professional Archetypes™ and Professional Classifications™.
• Generating profile recommendations.
• Rewriting LinkedIn headlines and About sections.
• Generating 30-day and 90-day action plans.
• Creating SWOT analysis.
• Identifying missing profile elements.
• Suggesting content themes.
• Summarizing submitted profile data.
• Identifying trust signals and visibility gaps.
• Generating social share copy and report summaries.
• Generating educational guidance.
• Supporting customer-service triage.

AI may assist with these outputs only when the user has provided sufficient data or when the system clearly discloses the limits of the analysis.

5. Prohibited AI Use Cases

• Making hiring or employment eligibility decisions.
• Making credit, lending, housing, insurance, legal, medical, education, or government-benefit decisions.
• Inferring protected characteristics.
• Ranking people by worth or social value.
• Impersonating a human reviewer without disclosure.
• Fabricating credentials, employment history, recommendations, or achievements.
• Generating fake testimonials or fake LinkedIn engagement.
• Scraping private user data without permission.
• Bypassing LinkedIn terms or platform protections.
• Promising algorithmic success.
• Claiming professional certification where none exists.
• Generating undisclosed fallback reports that appear identical to AI-generated reports.
• Producing legal, medical, financial, immigration, or tax advice.
• Creating manipulative dark-pattern marketing, false scarcity, false guarantees, or deceptive earnings claims.

6. AI Disclosure Standard

Every AI-generated or AI-assisted report must include an AI disclosure.

7. Data Collection Policy

7.1 Professional Influence Assessment™ Data

• Name, email, role or title, industry, assessment responses, generated score, category scores, Professional Classification™, Professional Archetype™, report history, credential ID, purchase information, report generation metadata.

7.2 Professional Influence Audit™ Data

• LinkedIn URL, headline, About section, featured items, skills, recent posts, photo status, banner status, creator mode status, recommendations count, followers, connections, optional self-assessment context.

10. AI Input Handling

• The system must not send unnecessary sensitive data to AI models.
• Remove unrelated personal information.
• Avoid including payment details or government ID data.
• Restrict inputs to the fields required for the report.
• Log prompt version and report version.
• For the $79.95 audit, the system should distinguish the AI-assigned audit score from the $9.95 self-assessment score because they measure different things.

12. Fallback Policy

Fallback reports must be disclosed. A deterministic fallback report should not be indistinguishable from a full AI-generated report.

Fallback reports must not pretend to be AI-generated, fabricate section-specific rewrites, create detailed analysis from missing data, assign high confidence if data is sparse, or include unsupported conclusions.

18. Hallucination Prevention Policy

AI must not invent employment history, education, credentials, awards, client results, follower counts, revenue, testimonials, endorsements, case studies, partnerships, media appearances, compliance status, verified identity, LinkedIn algorithm facts, benchmark claims, or percentile rankings.

If data is missing, AI must state that it was not provided.

19. Marketing Claims Policy

Approved Marketing Claims

• Discover your Professional Influence Score™.
• Receive AI-assisted LinkedIn profile recommendations based on the information you submit.
• Identify gaps in visibility, credibility, positioning, and trust signals.
• Get a personalized 30-day and 90-day improvement roadmap.
• Understand how your LinkedIn profile communicates your professional value.

Prohibited Marketing Claims

• Guaranteed to increase your LinkedIn reach.
• Guaranteed to get recruiter attention.
• Guaranteed to generate clients.
• Scientifically proven to measure influence.
• AI guarantees your profile will rank higher.

23. Human Review and Escalation Policy

Human review is required when a user disputes a score, requests refund or re-audit, reports an inaccurate output, alleges bias or discrimination, generated content may misrepresent credentials, legal or verification questions arise, fallback report was generated for a paid user, or the user requests deletion/export of data.

24. Refund and Re-Audit Policy

A user may be eligible for a complimentary re-audit if AI generation failed, fallback mode was used, report validation failed, report omitted required sections, the user accidentally submitted incomplete data and requests correction within 24 hours, or a technical issue prevented report delivery.

29. Security Policy

AI systems must not expose API keys, prompts containing secrets, internal scoring formulas, payment data, private customer records, identity-verification records, admin credentials, or raw AI logs to unauthorized users.

• Access control and environment variable management.
• Encryption in transit and at rest where feasible.
• Audit logs, least-privilege access, rate limiting, and abuse detection.

31. User Rights Policy

• Request access to their report.
• Correct submitted information.
• Request deletion.
• Request report regeneration after technical error.
• Opt out of marketing emails.
• Request badge deactivation.
• Request explanation of scores.
• Contact support.

36. Children and Minors Policy

The product is designed for professionals and should not knowingly collect data from children under 13. For users under 18, the product should not make career-impacting claims or collect unnecessary sensitive data.

41. Legal Disclaimer Policy

Legal disclaimers must be hardcoded, version-controlled, and reviewed. AI must not generate legal disclaimers dynamically.