RAI Certification Beta
The world’s first independent, accredited certification program of its kind.
Developed under the Global AI Action Alliance for the World Economic Forum (WEF), along with a diverse community of leading experts, RAI certification is based on objective assessments of fairness, bias, explainability, and other concrete metrics of responsibly built AI systems.
Why RAI Certification?
RAI Certification Allows for:
Recognizing, preparing for, and mitigating potential harmful effects of data and algorithm bias from blackbox AI.
Improving explainability to build customer trust and build competitive advantage with your AI systems.
Ensuring legal compliance and ethical alignment of your AI with your values and improve overall AI governance.
Addressing stakeholder expectations towards your company’s behavior and corporate social responsibility.
Avoiding reputational damage by increasing the transparency and explainability of AI decision making processes.
Leveraging our RAI community to solve AI trust problems and improve the performance and quality of your AI systems.
RAI Certification Beta areas of focus
- Fair Lending
- Fraud Detection
- Automated Diagnosis and Treatment
- Health Recommendation Systems
- Automated Human Resources
Join a RAI Certification Working Group
Recognizing that a project of this magnitude needs to be built by the community for the benefit of the community, we launched the RAI Certification Working Group December 2020 with WEF and SRI. Following the kick-off, we had an outpouring of support and interest in this work which led to numerous discussions with technical, data, governance, ethics, and industry experts. As such, we’ve created not one, but five working groups based on our areas of focus: Fair Lending, Fraud Detection, Automated Diagnosis and Treatment, and Automated Hiring.