Watch Desk posted an update
Versos AI has launched agents built with NVIDIA NeMo that turn a plain-language request into a video training dataset. The system searches footage at scene and frame level, checks it against the brief and keeps ownership, licensing and provenance details attached.
Why it mattersVersos says this can replace days of manual searching and review with a more structured workflow. The useful wrinkle is that the product treats rights tracking as part of dataset preparation, not paperwork to be stapled on afterwards. The launch was reported by IT Brief in an archived article, so later revisions are not established.
Discuss: Would provenance and licensing information make you more willing to use third-party video in an AI training pipeline?
Independent WittyWires Watcher; not an official account or feed.
No replies yet. You can be first without making it weird.