It's 2025, so it should be no surprise that another organization has sold its soul (entered into a licensing deal with an AI company) for an undisclosed sum. A new partnership allows UK-based Synthesia to access Shutterstock's content library for training its latest AI model, EXPRESS-2. This deal isn't the first of its kind for Shutterstock, which previously teamed up with OpenAI to sell stock images made using AI generator DALL-E 2. Synthesia creates avatars for corporate videos about topics such as cybersecurity and good communication at work. It aims to use Shutterstock's video data to "try out new approaches that will improve the performance of EXPRESS-2, and increase the realism and expressiveness of our AI generated avatars, bringing them closer to human-like performances.," Synthesia stated in a release. Typically, Synthesia uses actors to create avatars, paying to use their likeness for three years.

Video content creators need efficient tools to repurpose content, a task that often requires complex manual or automated searches. Crafting a new video from large video libraries remains a challenge. In this paper we introduce the task of Video Library Question Answering (VLQA) through an interoperable architecture that applies Retrieval Augmented Generation (RAG) to video libraries. We propose a system that uses large language models (LLMs) to generate search queries, retrieving relevant video moments indexed by speech and visual metadata. An answer generation module then integrates user queries with this metadata to produce responses with specific video timestamps. This approach shows promise in multimedia content retrieval, and AI-assisted video content creation.