Significant advances in generative AI in recent years have made artificial intelligence a top priority for businesses globally. As a result, large language models (LLMs) have become foundational in powering everything from virtual service agents to online search engines to fraud detection.
In entertainment media, LLMs will be foundational in powering rich search and discovery experiences, but they can’t do it alone. Because LLMs are prediction engines, they require complementary technologies to fact check the results they provide. These technologies improve accuracy, provide contextual relevance, enrich results, and align LLM outputs with real-world knowledge.
The Model Context Protocol (MCP) is ideal for ensuring that an LLM’s output is a reliable single source of truth facilitating a dynamic connection between an LLM and Gracenote’s knowledge base. This white paper details how MCP facilitates that connection to ensure that search and discovery experiences are rich and personalized, as well as accurate, recent and complete.
How RAG and MCP differ in powering AI-driven TV experiences
RAG and MCP each address the limitations of LLMs, but they approach the issue in fundamentally different ways.
Missing metadata in FAST programming may be hindering ad revenue
FAST channels will become increasingly dependent on metadata to inform ad buys in programmatic systems.
Streaming has viewers captive in a fractured CTV landscape
Streaming viewers have become overwhelmed by choice and fragmentation. This sentiment is mounting, and it has a range of downstream effects.