# AI Safety And Learning Disclaimer | Legal | SwaVid

Important limitations and safe-use rules for SwaVid AI-powered learning features.

Canonical: https://www.swavid.com/legal/ai-safety-disclaimer

Source: https://swavid.com/legal/ai-safety-disclaimer

# AI Safety And Learning Disclaimer

## Table Of Contents

## Educational Use Only

## No Professional Diagnosis

## Human Review

## Safety Controls

## AI System Classification And Educational Risk

## AI Output Duties And User Reliance

## Prohibited AI Uses

## Human Oversight, Contestability, And Reporting

## Related Legal Documents

Legal document | Last updated 20 June 2026

SwaVid uses AI to support learning, not to replace parents, teachers, schools, doctors, psychologists, counsellors, or qualified professionals.

SwaVid AI features generate educational explanations, practice prompts, diagnostic insights, learning plans, voice or chat responses, and parent or school reports. These outputs are educational aids and may be wrong, incomplete, or unsuitable.

SwaVid does not diagnose learning disabilities, psychological conditions, medical issues, attention disorders, behavioural disorders, mental-health conditions, or clinical needs. Consult qualified professionals for clinical, medical, therapeutic, or psychological decisions.

Parents, guardians, teachers, and schools should review important AI outputs before acting on them. SwaVid should not be the sole source for high-stakes academic, disciplinary, health, psychological, or placement decisions.

SwaVid AI features may include conversational tutoring, diagnostic analysis, adaptive learning recommendations, report generation, voice or speech processing, content generation, question generation, and analytics. Certain jurisdictions may classify education-related AI systems that evaluate learning outcomes, influence access to education, or steer learning processes as higher risk or subject to specific transparency, oversight, documentation, security, and quality obligations.

SwaVid therefore treats education AI as a controlled product area. Material AI features should be reviewed for intended purpose, user group, child impact, training or inference data, prompt inputs, output limitations, fallback behaviour, hallucination risk, harmful-content risk, bias risk, privacy risk, cybersecurity risk, and human oversight requirements.

Users must understand that AI outputs are probabilistic and may produce fluent but incorrect statements. A correct-looking explanation may contain factual errors. A generated report may overstate confidence. A tutoring answer may fail to detect a misconception. A generated question may not match the exact curriculum. A voice transcript may mishear a student. A recommendation may not account for emotional, medical, family, school, or disability factors.

Parents, guardians, teachers, schools, and students must not use SwaVid AI as the sole basis for medical, psychological, disciplinary, placement, admission, exam, disability, counselling, or high-stakes academic decisions. Human review is mandatory for important decisions.

SwaVid must not be used to generate abusive child content, exploitative content, harmful instructions, exam cheating content, discriminatory profiling, surveillance dossiers, psychological diagnosis, medical advice, self-harm encouragement, illegal conduct instructions, credential theft, malware, or content that violates school policy or applicable law.

Users must not attempt prompt injection, jailbreaks, system-prompt extraction, model inversion, data exfiltration, or manipulation of AI safety systems. SwaVid may block, log, restrict, or report such attempts where legally permitted.

SwaVid aims to keep parents, guardians, teachers, and schools in the review loop for important learning decisions. Users may report unsafe, incorrect, biased, inappropriate, or concerning AI outputs to info@swavid.com. Reports should include the account, student profile if relevant, date, feature, screenshot or text where available, and the reason for concern.

Where applicable law grants a right to human review, objection to profiling, explanation, rectification, or contesting automated processing, SwaVid will process the request through the privacy and grievance workflow.

The following documents form part of SwaVid's public legal framework and may be incorporated by reference where applicable. Formal notices and requests should be sent to info@swavid.com .

- 1 . Educational Use Only
- 2 . No Professional Diagnosis
- 3 . Human Review
- 4 . Safety Controls
- 5 . AI System Classification And Educational Risk
- 6 . AI Output Duties And User Reliance
- 7 . Prohibited AI Uses
- 8 . Human Oversight, Contestability, And Reporting
- ( a ) Do not enter private, unsafe, illegal, abusive, exploitative, or unrelated personal content into AI features.
- ( b ) Do not ask SwaVid to help cheat, bypass rules, harm others, or generate unsafe content.
- ( c ) Report concerning AI output, child-safety issues, or privacy concerns to info@swavid.com.
- Privacy Policy
- Terms and Conditions
- Children Privacy
- Cookie Policy
- Parent Consent
- Refunds
- Grievance
- Sub-processors
- Security
- DPA
- Student Data
- Data Policy

## Key Links

- [SwaVid](https://swavid.com/)
- [Legal](https://swavid.com/legal/privacy-policy)
- [info@swavid.com](https://swavid.commailto:info@swavid.com)
- [Privacy Policy](https://swavid.com/legal/privacy-policy)
- [Terms and Conditions](https://swavid.com/legal/terms-of-service)
- [Children Privacy](https://swavid.com/legal/children-privacy-notice)
- [Cookie Policy](https://swavid.com/legal/cookie-policy)
- [Parent Consent](https://swavid.com/legal/parent-consent-notice)
- [Refunds](https://swavid.com/legal/refund-cancellation-policy)
- [Grievance](https://swavid.com/legal/grievance-redressal)
- [Sub-processors](https://swavid.com/legal/sub-processors)
- [Security](https://swavid.com/legal/security)
- [DPA](https://swavid.com/legal/data-processing-addendum)
- [Student Data](https://swavid.com/legal/student-data-protection-agreement)
- [Data Policy](https://swavid.com/legal/data-policy)