Frequently Asked Questions

Generative AI systems can be misused for creating deepfakes, disinformation, and other manipulated content, which can have severe negative consequences for individuals and society. The challenge aims to surface solutions that ensure the public can trust the information they consume.

Anyone with an idea on policy direction and the ability to implement that policy direction in a technical approach can participate. Teams are encouraged to be multi-disciplinary, bringing together diverse skills from digital technology, public policy, and specific domain expertise.

The Global Trust Challenge consists of three phases:

• Phase 1: Proposal of Novel Policy Approaches and Technological Solutions.
• Phase 2: Creation and Testing of Prototypes.
• Phase 3: Live Pilots and Strategy for Scaling.

Submissions will be evaluated by a jury of experts from leading organizations in policy, technology, media, advocacy, and standards. Evaluation criteria are outlined in the challenge guidelines and will include the relevance and integration of proposals, alignment with standards for trustworthy AI, feasibility (both technical and policy), innovation and impact, and risk management. As teams progress through the phases, additional criteria such as usability, user feedback, resource efficiency, scalability, and community engagement will be considered. Rest assured, the evaluation process will be transparent and straightforward.

Teams may receive access to technical and policy experts, resources such as sandbox environments, data storage, compute capacity, and financial support for developing prototypes and pilots.

Teams must submit detailed proposals covering policy approaches, technological solutions, testing and validation plans, and an implementation strategy.

Participants must comply with confidentiality requirements, intellectual property management, conflict of interest disclosures, and terms and conditions of the challenge.

Teams must register by completing the registration form on the Global Challenge website: https://globalchallenge.ai/. For additional information, teams can contact the Global Trust Challenge Team at [email protected] .

The official communication language of the challenge is English, and the official currency is US dollars (US$). Proposals may be submitted in other languages, with a brief summary in English for the initial level screening. If selected, the challenge will leverage translation tools and global experts for evaluating the more detailed proposals.

Any changes to the guidelines will be made at the sole discretion of the organizing parties and communicated to all participating teams in a timely manner.