At Pixy we are committed to ensuring the responsible and ethical use of our platform, aligning with our privacy policy and safeguarding principles. This policy outlines the moderation framework for controlling the generation, sharing, or access to prohibited material, ensuring compliance with applicable laws and ethical standards.

The following types of content are explicitly prohibited on our platform:

Content that replicates, impersonates, or resembles real people, living or deceased, without explicit consent.

Material that promotes violence, discrimination, hate speech, or harassment against any individual or group based on race, ethnicity, religion, gender, sexual orientation, or other characteristics.

Sexually explicit material, content involving minors, or other exploitative media is strictly forbidden.

Material that violates local, national, or international laws, including fraud, misinformation, or content that incites illegal activities.

Generated content designed to mislead or deceive, including deepfakes or fabricated information intended to harm reputations or incite conflict. 

The following categories and examples represent prohibited words and phrases on the Pixy platform. This list is designed to ensure compliance with the company's ethical standards, applicable laws, and to maintain a safe and respectful environment. Note that this list is non-exhaustive and subject to updates as necessary. 

Words or phrases promoting hate, violence, or discrimination against individuals or groups based on race, ethnicity, religion, gender, sexual orientation, nationality, or disability: • Racial slurs or epithets • Anti-Semitic, Islamophobic, or other religious slurs • Derogatory terms related to gender or sexual orientation • Hate group slogans or acronyms 

Terms used to threaten, intimidate, or harass others, including incitement to violence:
• Explicit threats (e.g., "I will harm you")
• Derogatory insults or abusive language 
• Encouragement of self-harm or suicide 

Language related to sexually explicit material, exploitation of minors, or non-consensual content: 
• Terms referring to illegal sexual acts 
• Phrases promoting child exploitation 
• Explicit descriptions of sexual activity 

Language that incites violence, promotes harm, or glorifies dangerous acts:
• Instructions or encouragement to commit violent acts 
• Terms related to illegal weapons or explosives 
• Glorification of criminal activities 

Words or phrases associated with false claims, scams, or deceitful practices:
• Misleading health claims (e.g., "cure-all" for serious illnesses) 
• Financial scams (e.g., "Get rich quick")
• Fake news terminology (e.g., intentionally misleading political phrases)

Terms that encourage the generation of content resembling real individuals without consent:
• Phrases like "Create a celebrity look-alike" or "Replicate [famous person's name]" 
• "Generate realistic voice for [specific name]"

Language that references illicit activities, substances, or other legally prohibited actions:
• Drug-related terms (e.g., "Buy [illicit drug] online")
• Terms related to trafficking or smuggling 
• References to hacking or cybercrime

Phrases or words related to encouraging self-harm, suicide, or harm to others:
• "How to end my life"
• Encouragement to engage in risky behaviors

Language that encourages sharing or misuse of personal data without consent:
• "How to steal someone's identity" 
• Phrases targeting unauthorized data extraction (e.g., "Hack someone's email")

Phrases or terms prohibited under specific national or international laws, including defamation or promotion of banned ideologies: 
• References to banned political groups 
• Defamatory terms targeting individuals or entities 

‍To enforce these prohibitions:
• Filters are set to block prompts containing these terms.
• Content flagged for violating these guidelines will be reviewed and removed promptly.
• Users generating content containing prohibited language may face account suspension or termination. 

This list helps Pixy maintain a safe and respectful platform while complying with ethical and legal standards globally. Regular reviews will ensure the list remains updated to address emerging risks and concerns. 

• Input Filters:
All prompts are scanned for restricted keywords or patterns indicative of attempts to generate prohibited material.
• AI Safety Models: The generation engine incorporates ethical constraints to prevent the creation of harmful or restricted content. 

• Automated Scanning: Generated content is reviewed through AI-driven moderation tools to identify potential policy violations.
 • Human Review: Flagged outputs are subjected to additional scrutiny by trained content moderators. 

• Users must agree to our Terms of Service, which prohibit attempts to generate or disseminate restricted content.
• Users who violate the policy may face:
o Temporary suspension of access. 
o Permanent account termination.
o Legal consequences for severe or repeated violations. 

• Content Flagging: Users and moderators can flag potentially inappropriate content.
• Investigation and Action: Reports are investigated promptly, and appropriate action is taken, including content removal and user suspension if necessary. 
• Appeals Process: Users can appeal moderation decisions through our support team.

The layered approach of pre-generation safeguards and post-generation monitoring minimizes the likelihood of creating prohibited material. Input filters and safety models ensure that prompts likely to result in inappropriate outputs are blocked before content is generated.

Automated scanning combined with human oversight enables swift identification and action against violations. This ensures that harmful content is removed quickly, minimizing exposure and impact.

The enforcement of strong user accountability measures, such as suspension and legal consequences, acts as a deterrent against misuse. The clear terms outlined in the Privacy Policy reinforce responsible platform use. 

Providing mechanisms for users to flag content and appeal decisions fosters transparency and trust, ensuring fairness while maintaining strict adherence to platform policies. By integrating technological safeguards, human oversight, and user accountability, this policy effectively mitigates the generation and dissemination of prohibited material while promoting a safe and ethical platform environment.