Before this: What Are AI Assets and How to Use Them?
What Is Video to AI Video?
Video to AI Video is a separate AI feature whose input is a video, not a photo: it lives in Settings → AI Assets → Video to AI Video tab. The Photo to AI Video feature in Documents 20/21 (before 3.4.2 the tab was called AI Video) takes a photo as input and is a completely different feature; the two are not in the same tab, don't mix them up. Here, a short video the guest captures (or the operator uploads) is transformed by AI into a new video.
- Pro AI subscription or higher, the general restriction on the AI Assets page, which also applies to this tab
- An event must be selected/open
- Enough AI credit, every test and final production spends credit
- Internet connection, production runs server-side (Grok model)
The AI model in use (Grok) has an 8-second video-to-video input limit. If the video you select/upload exceeds this, the app automatically trims it to the first 8 seconds: it doesn't throw an error, it just trims silently. When Video to AI Video is chosen as the event type, the booth's own capture duration should be kept short to match; otherwise the end of the guest's captured video never makes it into the AI transformation.
Step by Step
Step 1 · Open the tab: library and overview
Settings → AI Assets → Video to AI Video
Open the AI Assets page from Settings; switch to the Video to AI Video tab in the tab bar at the top. Below the tabs, the description Restyle a recorded guest video into a new AI-generated video. appears. At the top there's a Surprise Effect toggle (when on, the guest doesn't pick an effect, a random video effect is applied). Below it is the Video to AI Video Effects Library heading and a grid of the existing video cards, with Browse Library and Create New buttons in the top right; if there is at least one video in the library, Remove All also appears to their left (confirmation: Remove all effects? → Delete; the videos are only removed from this event and stay in the library). Cards with a cover video have a round play button in the middle: tap this button to preview (on desktop, hovering over the card also plays it). Clicking anywhere else on a card, for a card you created yourself, it opens the Edit & Customize modal; for a card that isn't yours / is a default one, it opens the Clone & Customize modal (the same ownership rule as Photo to AI Video applies here too). Drag the cards to change their order; on a touchscreen, press and hold a card for about half a second to start dragging (a quick finger swipe scrolls the page).
The AI Video Fade Transition section below Surprise Effect sets how the original video and the AI result are combined when shown to the guest: Show the original recorded video, cross-fade into the AI result, then cross-fade back to the original. The handles set where each transition starts; Transition Duration sets how long each cross-fade lasts. Drag both handles to 0 for a direct cut with no fade. The labels at the top show the selected moments (To AI at, Back to original at); when the transition is off, it reads Off, direct cut.
🔎 On-screen elements: Surprise Effect toggle; library heading + counter; Remove All (when there are videos) / Browse Library / Create New buttons; grid of video cards (round play button in the middle; on desktop, plays on hover over the cover)

Step 2 · Browse Libraryadd ready-made videos to the event
Video to AI Video → Browse Library
This modal is split into 4 tabs: Editor Video Effects (editor/system catalog), My Video Effects (the ones you created), My Favorites, Default Ai Videos. Clicking a card selects/deselects it; cards already added to the event show an "Added" badge (this badge doesn't have a separate translated string, it stays in English). Once you've made your selection, the button at the bottom of the modal adds the selected ones to the event's library.
🔎 On-screen elements: 4-tab bar at the top; card grid with a checkmark icon on selected cards; an English Added badge on cards already added

Step 3 · Create Newinput video and test
Video to AI Video → Create a Video to AI Video Effect
The modal that opens when you press Create New shows a Video to AI Video badge, the credit cost, and the estimated production time (~seconds) in its title. The left preview box is the input video: unlike AI Video, this isn't a photo, it's an actual video file; the Upload Your Own Video button below it lets you pick your own video file from the native file picker (it plays with sound on hover/tap; while the video you picked is being prepared and uploaded, the box shows Loading). The right box is where the test result (the generated AI video) appears. When the test finishes, the result also opens automatically in a full-screen preview window (Download / Close); the eye icon (Preview) at the top right of the result box reopens this window (3.5.1 and later).
🔎 On-screen elements: Video to AI Video badge + credit + duration badges; input video + Upload Your Own Video button in the left box; Effect output placeholder in the right box when empty, eye (Preview) and download icons after the test

Step 4 · Effect name, Prompt, and personalization chips
Create a Video to AI Video Effect, form
Fill in the Effect Name (required, e.g. Neon Cyberpunk) and Prompt (required, the text describing how the video should be transformed) fields. Below the Prompt box, 7 chips are listed under the Available personalization variables: label: {Name} {Number} {Color} {Interest} {Scene} {Weather} {CustomPrompt}. Clicking a chip pastes that variable at the cursor position; at capture time, the server replaces these placeholders with real session data (the exact same mechanism as the chips in the AI Effect / AI Video creation modals). When you add a variable to the prompt, the Test values panel appears below the chips (Only used for Test, at the booth your guests enter these values.); only the fields used in the prompt are pre-filled with sample values, and Test Effect runs with these values (3.4.0 and later).
🔎 On-screen elements: Effect Name text box; Prompt multi-line text box; 7 chip buttons below it ({Name}…{CustomPrompt}); if the prompt contains a variable, the Test values panel

Step 5 · Cover media and save
Create a Video to AI Video Effect, Cover Media
In the Cover Media section, you can pick an image or video with the Pick cover image or video button (recommended 640x400px). If you don't choose a cover, the result video from a successful Test Effect can be made the cover with the Use as Cover button. Once the form is complete, press the Save Effect / Save Changes / Save as New Effect button at the bottom (depending on the mode); the Make It Public option is hidden for this feature (it comes off and can't be changed).
🔎 On-screen elements: Cover Media preview box + Pick cover image or video button; save button in the bottom bar

Step 6 · Selecting it as an event type
Event creation wizard → Step 7: Type selection
To use these effects live in the booth, check the Video to AI Video card in the event type selection (Document 09, Step 7, now one of 12 types). The booth captures a video from the guest, and the captured video is transformed into an AI video using one of the effects defined in this tab (or a random one if Surprise Effect is on). Remember: if the input video is longer than 8 seconds, it's automatically trimmed (see the warning above).
To customize the effect selection screen the guest sees: set the title in the Screen Title field of the Video to AI Video section on the Screen Design → Texts tab, and the background in the Video to AI Video Screen field of the Custom Backgrounds list on the Screen Design → General tab (if no background is set, the general default background is used).
3.5.3 and later: the title and background of this screen are set separately from those of the Photo to AI Video screen. In earlier versions the title was shared with the field in the AI Video Page section, and there was no background field for this screen. If the new title field is left empty, the old shared title continues to be used.
🔎 On-screen elements: The Video to AI Video card in the event type selection screen (with a video preview)

Photo to AI Video (Documents 20/21; before 3.4.2 the tab was called 'AI Video'): input is a photo, output is a video. Video to AI Video (this document): input is a video (up to 8 sec), output is a video. The two live in separate tabs, separate libraries, and separate event-type cards, don't look for one effect in the other's library, you won't find it.
Related topics
AI Assets Overview How to Create Custom AI Video?