What the AI can and cannot do
A plain list: what works now, what needs signing in, and what this site will not pretend to do. Including the boundaries — no live-action generation, no quality restoration, no watermark removal.
Every AI tool is better at listing what it can do than what it cannot, which is how people end up disappointed halfway through a project.
Editing, subtitles, transitions, effects, character animation, scene building and full video creation from a description all work now, in the browser, without an account.
What it does not do is written down too: no photographic live-action generation, no restoring quality that was never recorded, no stripping someone else's watermark. It is here rather than discovered halfway through a project.
FAQ
Can it generate realistic video of real places?
No. It composes from drawn scenes, characters, text and shapes. Photographic generation is not something this site offers.
Can it make a blurry video sharp?
No. Enlarging cannot invent detail that was never recorded, and this site does not claim otherwise.
Can it take a mark off a video?
No, and it is not something we will add.
What needs an account?
Four things: AI subtitles, photoreal asset generation, 4K cloud rendering and cloud storage. Each has a daily free allowance.
What works with no account at all?
Everything else — all 72 tools, AI editing, and AI video creation.
Does any of this send my video to a server?
No. The assistant works through the project's structure — what clips exist, where they sit, what they say — rather than by uploading footage. The four features that do need a server are named above, and none of them sends your video.