Built with security, ethics, and safety at its core
Our AI video platform is designed with security and ethics first, embedding protection into infrastructure and policies while enabling innovation. Powerful technology must prioritize trust, not treat it as optional.
Comprehensive security for every frame
Our commitment extends beyond basic security measures; we implement comprehensive safeguards at every step of the video generation process. From the moment you upload your content to the final delivery of your AI-generated videos, your data is protected by industry-leading security protocols. We continuously evolve our security practices to stay ahead of emerging threats while maintaining the seamless, user-friendly experience that makes HeyGen the trusted choice for AI video creation.
GDPR
SOC 2 TYPE II
CCPA
AI ACT
DPFCertified to meet global security and compliance standards
Core values
Our core values establish a foundation of trust, responsibility, and continuous improvement for the Trust and Safety team.
Integrity
We uphold the highest standards of honesty and transparency, ensuring that all decisions and actions align with ethical principles and our commitment to user safety.
Collaboration
We work closely with users, partners, and industry experts to create a safer digital ecosystem, leveraging collective knowledge and feedback to improve our policies and practices.
Accountability
We take full responsibility for our policies, decisions, and actions, striving to be reliable and trustworthy stewards of our platform's integrity.
Proactivity
We anticipate and mitigate potential risks by constantly monitoring, analyzing, and improving our safety frameworks, addressing issues before they escalate.
User-centric
The safety and well-being of our users are at the heart of everything we do. We actively listen to their concerns and prioritize their protection in every policy and decision.
Respect for privacy
We safeguard user data with the utmost care, ensuring that privacy is protected and respected throughout every interaction on our platform.
Innovation with responsibility
We embrace cutting-edge technology but are committed to implementing it thoughtfully, ensuring that safety measures evolve alongside innovation.
Key principles
Ownership and consent
You retain full ownership of any avatars you create—whether photo or custom—but you must ensure you have the legal rights and explicit consent of any individual whose likeness is used. This includes honoring removal requests from those depicted.
Strict prohibitions
We strictly forbid content that is violent, hateful, deceptive, sexually explicit, infringing, abusive, political (e.g., campaigning), or harmful to minors. This also covers disallowed content such as scams, misinformation, or restricted goods.
Conditional allowances
Certain sensitive themes—like educational material containing mature elements or respectful religious content—may be allowed but only if clearly labeled and used appropriately.
Enforcement and review process
At HeyGen, we believe that revolutionary AI technology must come with the highest standards of security and ethics embedded from the start. Our dedicated Trust & Safety team ensures your data is secure and our AI is used ethically.
Automated moderation
We use ML-powered scanning tools to detect potentially non-permissible content as it's uploaded or generated.
Human oversight
Flagged or critical content undergoes manual review by trained moderators to ensure fairness and accuracy.
Action and appeals
Enforcement measures
If a violation occurs, HeyGen may remove the content, block its generation, suspend or terminate access, or report to authorities as appropriate.
Appeal process
Users can appeal moderation decisions by emailing [email protected], including details of the flagged content and reasoning. Appeals are reviewed—ideally by a different moderator than the first—but submission does not guarantee a response.
Policy evolution
Our moderation policies evolve with emerging threats, user feedback, and societal changes. Updates take effect immediately upon posting—they're part of our ongoing commitment to trust and safety.
Continuous monitoring
App security
Data security
Network security
Infrastructure security
Organization security
Product security
Subprocessors
Microsoft Azure
Text-to-speech
Data location: United States
Intercom
Customer support and help center
Data location: United States
Amazon Web Services
Web hosting, data storage, image content moderation, and video processing
Data location: United States
Datadog
Application monitoring and alerting, log storage and collection
Data location: United States
ElevenLabs
Text-to-speech
Data location: United States
Cloudflare
DNS
Data location: United States
Have questions? We have answers.
What does Trust and Safety mean at HeyGen?
At HeyGen, trust and safety means building our AI video platform with ethics in AI and user protection at its core. We prioritize integrity, accountability, and respect for privacy in every decision, ensuring that innovation is balanced with responsibility. By addressing AI safety concerns directly and embedding protections into our infrastructure and policies, we foster a respectful, inclusive environment where creativity can thrive without compromising security.
How does HeyGen ensure compliance with SOC 2 Type 2 and GDPR standards?
HeyGen meets rigorous compliance requirements for global standards such as SOC 2 Type 2, GDPR, CCPA, the Data Privacy Framework, and the EU AI Act. These certifications and attestations demonstrate that we address AI safety risks alongside data protection obligations. Regular audits, documentation, and transparent reporting ensure our systems meet the highest levels of security and privacy while aligning with international regulatory frameworks.
What steps does HeyGen take to protect user privacy and data security?
Protecting privacy is central to our approach to AI safety. Every step of the video generation process—from upload to final delivery—is safeguarded with industry-leading security protocols. We use encryption, proactive monitoring, and evolving best practices from AI safety research to anticipate and defend against threats. Beyond technical measures, we uphold strict values of privacy and ethical responsibility to ensure data is handled with the utmost care.
How does HeyGen enforce policies to keep its platform safe and compliant?
HeyGen enforces a robust Acceptable Use and Moderation Policy that proactively addresses AI safety concerns and mitigates AI safety risks. Our safeguards include:
- Ownership and consent: Users must have legal rights and explicit consent for any likeness they upload.
- Strict prohibitions: Content that is violent, hateful, deceptive, sexually explicit, infringing, political, or harmful to minors is strictly forbidden.
- Conditional allowances: Sensitive topics may be allowed when used responsibly for education or respectful purposes.
- Enforcement and review: Machine-learning tools detect potential violations, followed by human moderator review to ensure fairness.
- Appeals process: Users can appeal moderation decisions for an additional review.
By evolving our policies with new risks and community feedback, we keep HeyGen aligned with best practices in AI safety and ethical AI use.
What ethical AI principles guide HeyGen’s technology development?
HeyGen’s technology development is grounded in ethics in AI and guided by principles that ensure safety and responsibility:
- Integrity: Transparency and honesty in how we design and deploy AI.
- Accountability: Taking responsibility for policies, outcomes, and the safe use of AI.
- User-centric design: Prioritizing well-being and addressing AI safety concerns raised by users.
- Responsible innovation: Advancing AI while actively managing AI safety risks.
- Collaboration: Working with partners, experts, and the broader community to share knowledge from AI safety research.
- Proactivity: Anticipating challenges to prevent misuse or harm.
- Respect for privacy: Safeguarding data to maintain user trust.
These values keep AI safety and ethics at the forefront of everything we build, ensuring our platform remains trustworthy, secure, and responsible.
Does HeyGen train its model on my data?
No. We ensure that our enterprise customers' data is not used to train our AI systems, nor is it combined with any other data for this purpose.
Non-enterprise customers may be used to train and enhance the models that power our Services. Users may opt-out of this training, see: https://www.heygen.com/privacy.
Does HeyGen’s subprocessors train its model on my business data?
No. HeyGen has agreements in place with our AI vendors that do not allow training of their models using HeyGen customer data.
Start creating videos with AI
See how businesses like yours scale content creation and drive growth with the most innovative AI video.
