Aggregation Alone Lacks a Moat
Velvet
Aggregation is a good onboarding wedge, but it is a weak long term moat in AI video because the same model vendors and features are quickly becoming available inside larger editing and hosting platforms. Velvet is strongest when it turns scattered tools into a finished workflow for one job, like shipping product launch videos, but the durable value sits in owning templates, brand context, review loops, publishing, and results after generation, not just model access.
-
The market is already moving from single feature AI tools toward full workflow suites. Synthesia bundled avatar creation, screen recording, editing, translation, hosting, analytics, and lead capture, while Descript owns capture through editing. That is how AI video products become system of record instead of temporary generators.
-
Horizontal products can copy aggregation fastest because they already own distribution. Canva plugs in third party AI features for a huge installed base, Descript starts at script and editing, and CapCut can absorb casual creators with free mobile editing before they ever need a paid specialist tool.
-
The clearest defense is vertical depth. Velvet is positioned around product launch videos, and adjacent winners in AI video keep differentiating by going deeper into a use case, customer segment, or workflow, like enterprise training, sales outreach, or social clip repurposing, rather than by being a neutral model switchboard.
From here, the winning shape is likely a narrower but deeper platform. Velvet can expand TAM by turning creation into an ongoing operating loop, with reusable launch templates, asset libraries, approvals, hosting, and distribution tied to product marketing teams. That is the path from convenient aggregator to durable workflow software.