Details
CCI is built on Amazon Bedrock and Anthropic's Claude 3.5, using a retrieval-augmented generation (RAG) workflow to create a unified knowledge base from hundreds of documents and large statistical databases covering rider profiles, race times, and event histories. Commentators query it using natural language during live broadcasts. The system also uses Amazon Textract for document processing, Amazon Translate for multilingual content, and Amazon Comprehend for information organization. Prior to CCI, commentators spent hours researching across multiple sources before and during races, a manual process AWS explicitly confirmed.
Have evidence about Warner Bros. Discovery's AI practices? Submit a report.
Report a Sighting →