Acceptable Use Policy

AI Models Underlying Our Service

EduForge is an independent platform that uses AI models from third-party providers to generate training content. We are not affiliated with, endorsed by, or an official representative of any AI model provider.

Underlying AI models used by EduForge:

• DeepSeek AI — Primary model for course generation, PPT creation, quiz generation, and AI practice coaching

Content generated by EduForge is produced by DeepSeek AI models. We do not guarantee specific output quality and are not responsible for content generated by third-party AI models.

1. Purpose and Scope

This Acceptable Use Policy ("AUP") outlines the rules, guidelines, and prohibitions for using EduForge's AI-powered training content generation services ("Service"). This policy applies to all users, including individuals, teams, and organizations, regardless of subscription tier.

By using the Service, you agree to comply with this AUP. Violations may result in enforcement actions as described in Section 6.

2. Prohibited Content Categories

You may NOT use EduForge to generate, create, upload, store, transmit, or disseminate any of the following:

2.1 Child Sexual Abuse Material (CSAM)

ZERO TOLERANCE. Any content depicting, describing, or simulating child sexual abuse is strictly prohibited. This includes:

• Any visual depiction of a minor engaged in sexually explicit conduct
• Any content that sexualizes minors in any way
• Any AI-generated or synthesized content depicting minors in sexual contexts
• Any links or references to external sources containing CSAM

All violations result in immediate account termination and mandatory reporting to law enforcement.

2.2 Deepfake and Synthetic Media

You may NOT generate, disseminate, or assist in creating:

• Deepfake audio, video, or images impersonating real individuals without their consent
• Synthetic media intended to mislead viewers about its source or authenticity
• AI-synthesized content impersonating public figures, celebrities, or private individuals
• Fraudulent or deceptive synthetic media intended for financial gain, manipulation, or defamation
• Non-consensual intimate imagery (synthetic or real)

2.3 Hate Speech and Discrimination

• Content promoting hatred, violence, or discrimination against protected groups based on race, ethnicity, religion, gender, sexual orientation, disability, or national origin
• Content that harasses, intimidates, or threatens specific individuals or groups
• White supremacist, extremist, or terrorist ideology and propaganda

2.4 Violent and Criminal Content

• Instructions for creating weapons, explosives, or harmful devices
• Content promoting, encouraging, or depicting terrorism or violent extremism
• Content facilitating human trafficking, slavery, or forced labor
• Content related to drug manufacturing or illegal drug use instructions
• Content that could be used to commit fraud, identity theft, or financial crimes

2.5 Malicious and Harmful Software

• Malware, viruses, ransomware, or any malicious code
• Phishing content or social engineering templates
• Content designed to exploit vulnerabilities or compromise systems
• Hacking tools, unauthorized system access mechanisms, or credential harvesting tools

2.6 Intellectual Property and Privacy Violations

• Content infringing on copyrights, trademarks, or patents
• Unauthorized use of proprietary information or trade secrets
• Personal data of individuals shared without their consent (doxing)
• Content facilitating piracy or unauthorized distribution of copyrighted materials
• Training materials containing confidential or proprietary information without authorization

3. AI-Specific Prohibitions

In addition to the above categories, given the AI-powered nature of our Service:

• Model Training Prohibition: You shall NOT use the Service's inputs, outputs, model responses, or any derivative data to train, fine-tune, benchmark, knowledge-distill, or otherwise develop any AI/ML model that directly or indirectly competes with the Service or EduForge products
• Large-Scale Extraction: You shall NOT systematically crawl, extract, or collect model outputs on a large scale for purposes beyond your authorized personal use
• Professional Misrepresentation: In potentially harmful scenarios, you shall NOT pass off AI-generated content as the work of human professionals (such as practicing doctors, lawyers, or financial advisors)
• AI Service Disruption: You shall NOT attempt to manipulate, interfere with, or extract information from the underlying AI models beyond normal Service usage

4. Content Moderation System

EduForge employs a multi-layer content moderation system to detect and prevent prohibited content before and after generation.

4.1 Pre-Generation Scanning (Prompt Sift)

Before processing any user request, EduForge runs the input through Waffo Prompt Sift — a real-time content safety API that screens text prompts for potentially prohibited content, including CSAM-related requests, violence, hate speech, and other harmful intent. Requests flagged by Prompt Sift are blocked before they reach the AI model.

Additionally, EduForge maintains an internal prohibited keyword and pattern blocklist that is actively updated to cover emerging harmful content patterns.

4.2 Post-Generation Scanning

AI-generated content is reviewed before being presented to the user. Content that triggers safety filters is withheld, and the user receives a message indicating the content does not meet our acceptable use standards.

4.3 Human Review

Content flagged by automated systems enters a human review queue. All reports of policy violations are reviewed by trained Trust & Safety personnel. For L1 violations (CSAM, terrorism), reports are escalated to senior reviewers and may result in immediate law enforcement referral.

4.4 Moderation Audit Log

All moderation decisions — both automated and human — are logged. Logs are retained for a minimum of 12 months and may be disclosed to authorities in connection with verified violations.

5. Content Moderation Tiers

EduForge employs a tiered content moderation system based on violation severity:

Tier	Severity	Examples	Response Time
L1 - Critical	Immediate action required	CSAM, terrorism content, deepfake fraud	Within 1 hour
L2 - High	Urgent review needed	Hate speech, non-consensual deepfake, violence instructions	Within 4 hours
L3 - Medium	Review within business hours	Copyright infringement, spam content, minor policy violations	Within 24 hours
L4 - Low	Routine monitoring	Low-quality content, minor spam, ambiguous cases	Within 72 hours

6. Enforcement Actions

When violations are detected or reported, EduForge may take the following enforcement actions based on violation severity and user history:

Enforcement Action	Description	Applied When
Content Removal	Deletion of violating content from our systems	L1-L4 violations
Warning Notice	Formal written warning to user account	L3-L4 first offenses, accidental violations
Feature Suspension	Temporary disablement of specific features (e.g., AI generation)	L2-L3 repeated violations
Account Suspension	Temporary account lockout (7-30 days)	L2 repeated violations, L3 serious violations
Account Termination	Permanent account deletion and data purge	L1 violations, L2-L3 repeated serious violations
Law Enforcement Referral	Mandatory reporting and cooperation with authorities	CSAM, terrorism, fraud involving real victims
Legal Action	Civil litigation for damages or injunctive relief	Significant financial harm, repeated violations

7. User Notification

When enforcement actions are taken, users will receive:

• L1 actions: No prior notice for account termination; law enforcement may be notified first
• L2-L3 actions: Notice via email within 24 hours of action, including reason and evidence summary
• L4 actions: Notice via email or in-app notification within 72 hours

8. Appeals Mechanism

Users may appeal enforcement actions within 30 days of receiving notice. The appeal process:

• Step 1 - Internal Review: Submit appeal via appeals@eduforge.com with your account email, action reference number, and detailed explanation of why you believe the action was taken in error
• Step 2 - Independent Review: Appeals are reviewed by a team member not involved in the original decision
• Step 3 - Decision: We commit to responding within 10 business days; the decision is final for L1 violations
• Step 4 - External Escalation: If you remain unsatisfied, you may escalate to our dispute resolution partner or relevant consumer protection authority

9. Reporting Violations

If you become aware of any content or activity that violates this AUP, please report it immediately through one of the following channels:

Report Type	Contact	Expected Response
CSAM or Child Safety Concerns	safety@eduforge.com (24/7 priority)	Within 1 hour
Deepfake or Synthetic Media	deepfake@eduforge.com	Within 4 hours
General Policy Violation	abuse@eduforge.com	Within 24 hours
Security Vulnerabilities	security@eduforge.com	Within 24 hours
General Support	support@eduforge.com	Within 48 hours

Anonymous Reporting: While we encourage identifying yourself for follow-up questions, you may report anonymously. For L1 violations (CSAM, terrorism), anonymous reports will still be acted upon immediately.

10. Policy Updates

We may update this AUP from time to time to reflect changes in our Service, legal requirements, or industry standards. Material changes will be communicated:

• Via email to registered users at least 14 days in advance
• Via notice on our website at least 30 days in advance
• Via in-app notification upon next login

Continued use of the Service after changes constitutes acceptance of the updated AUP.

11. Contact

Questions about this Acceptable Use Policy can be directed to:

• Policy Inquiries: legal@eduforge.com
• Appeals: appeals@eduforge.com
• General Support: support@eduforge.com