Conference Presentation, Fireside Chat, Interview
Beyond Creativity: Why Generative Media Is Becoming Core Infrastructure | Fal | RAISE Summit 2026
- The generative media market is projected to surpass the LLM market in total scale, with video models expected to dominate the business future due to their higher inference difficulty and compute intensity.
- Revenue composition is shifting toward video generation, which currently accounts for 70% of income, although image editing is predicted to become the majority revenue driver given its faster growth rate over the past year.
- Video models are anticipated to eventually takeover the entirety of the education sector and telehealth operator interactions, with founders encouraged to target these verticals for significant opportunity.
- The global ecosystem requires substantial compute financing to avoid growth stagnation, leading to plans for facilitating resource access through renting or self-procurement.
- Professional workflows are expected to incur significant compute costs, reaching tens of thousands of dollars monthly for individual batch creation, with requirements rising as resolution targets 4k real-time movie streaming and durations increase.
- Image editing capabilities are forecast to continue improving for scale, while closed-loop advertising and human-in-the-loop use cases are expected to see significant advancements as agent performance and measurability improve.
- Future video model controllability and duration will rely on adding more context references, with labs actively developing features for consistent characters, cuts, and lighting to address current cinematic control limitations estimated at only 20% capability.
- Current state assessments place User Generated Content (UGC) video capabilities at 65-70% completion and music model capabilities at nearly 100%, driving expectations for IP holders to monetize existing content and recreate shows for fans.