From: redpointai
Integrity and safety are critical for HeyGen’s business, especially when serving large enterprise customers [00:00:04]. The company’s work on integrity focuses on two main areas: avatar creation and video content moderation [00:00:12].
Hybrid Moderation Approach
HeyGen employs a hybrid solution combining AI models and human review to ensure content adheres to its policies [00:00:58].
Consent for Avatars
To create an avatar on HeyGen, users are always required to provide video consent [00:00:20]. This process includes an advanced AI system that matches the consent footage to ensure it’s the same person, making it nearly impossible to create someone’s avatar without their explicit consent [00:00:33].
Content Review
Content created on the platform undergoes a dual review process:
- AI-driven review [00:00:35]
- Human moderation team review [00:00:38]
This combined approach ensures that all footage aligns with expectations [00:00:42].
Prohibited Content Categories
HeyGen’s platform moderation policy strictly prohibits several categories of content, including:
- Hate speech [00:00:46]
- Misinformation for children [00:00:48]
- Any political campaign or related information [00:00:54]