AI Content Multiplier

The AI Content Multiplier repurposes a source video in two modes:

• Highlights — the AI selects the most engaging moments and generates individual clips and compilations, optionally reframing them to a different aspect ratio.
• Reframing — the selected timeframe is preserved and reframed to a different aspect ratio; no highlights, clips, or compilations are generated. When you leave the default full timeframe unchanged, the output contains the entire source video.

In Highlights mode, clips are individual, straight cuts from the source video, while compilations are highlights-style videos that concatenate multiple individual clips. Generated clips and compilations can be previewed, downloaded, or ingested to become full-fledged Brightcove videos.

Setting up templates

Templates allow you to predefine a collection of settings that are reused whenever clips are generated.

Creating a new template

1. In the Admin module, go to AI Clips Generation.
2. Click New Template +

   

3. Configure your template settings and click Add.

   

Template settings

Generating AI Clips

1. In Video Cloud Studio, open the Media module.
2. Click on a video title to open the Video Details page.
3. Click the Generate AI Clips button in the top-right corner.

   

4. Configure the generation settings described below.
5. Click Generate to start clip generation.

   

Generation settings

Reviewing and managing AI Clips

Once generation is complete, you can preview and decide which clips and compilations to download or ingest.

1. Go to the Media module and select the source video.
2. Click Select AI Clips in the top-right corner.

   

Viewing clips and compilations

Use the menu in the top-left corner of the Select AI Clips modal to switch between All, Clips, and Compilation. You can also identify a compilation by the multiple timestamps shown on its card.

Sorting clips

By default, clips are ordered by Virality Score. Use the sort control at the top of the clip list to order clips by Virality Score, Liked, or Duration. The virality score estimates how engaging a clip is by evaluating various parameters, including hook strength, surprise, emotional intensity, whether the clip is a peak moment from the source video, narrative completeness (whether the clip works as a complete micro-narrative), and context independence (whether the content can be understood without knowing the broader context of the source video). Below each video preview, you can see a brief reason for the score.

Available clip actions

Each clip card and the bottom action bar expose the following actions:

• Preview — click a clip in the list to load it in the right-hand player.
• Like — click the heart icon on a clip card to mark it for future reference.
• Trim — with a clip selected, click the scissors icon above the preview player to trim the clip (see Trimming a clip below).
• Download — click the download icon above the preview player to save the clip to your device.
• Save Video — click Save Video below the preview to ingest the clip into your video's AI Clips section.
• Discard — click Discard below the preview to remove the clip from the selection. The bottom action bar also offers bulk Discard for multiple selected clips.
• Combine — select two or more clips, then click Combine in the bottom action bar to merge them into a single video (see Combining clips below).

Trimming a clip

1. Select a clip and click the scissors icon above the preview player.
2. Drag the trim handles on the timeline to set the new start and end points, then click Trim.

   

Combining clips

Select two or more clips, then click Combine in the bottom action bar. Enter a title for the combined video; a title is required before you can save it. To reorder clips in the combined video, drag and drop them using the six-dot handle before saving.

Generating more clips and compilations

To generate more clips and compilations for a video that already has generated output, close the Select AI Clips modal and click Confirm and Discard in the Unsaved Clips dialog. This discards all unsaved clips and compilations and lets you generate them again from the Generate AI Clips button.

Finding clips and compilations in the media library

By default, ingested clips and compilations are hidden in the Media table. Videos with saved or ingested clips or compilations display an expand arrow next to their thumbnail; click it to reveal the associated output. The AI Clips button above the table applies the Videos with AI Clips filter automatically.

Click the AI Clips filter chip to open the filter popup and select one of these options:

• Videos with AI Clips
• Videos without AI Clips
• Only AI Clips

Detaching clips

You can detach a clip from its parent video to create a standalone video. This will remove the clip's link to the original video, but it won't delete any content.

1. Click the three-dot menu for a clip.
2. Select Detach AI Clip from parent.
3. Confirm by clicking Detach Video

   

API access

The AI Content Multiplier is available via the Ingest API. All endpoints require OAuth with scope video-cloud/video/read, except where noted.

Create / Get / Update / Delete AI Content Multiplier job

Endpoint:

POST | GET | PATCH | DELETE https://ingest.api.brightcove.com/v1/accounts/{account_id}/videos/{video_id}/ai/content-multiplier
Scope: video-cloud/video/read

POST – Create clip generation job

Starts AI clip generation for a video. Request body (all fields optional except as noted):

• video_name: optional string. Used when combining multiple clips (e.g. video_id-1-3-5) as the name for the combined output.
• min_duration: optional integer. Minimum clip duration in seconds.
• max_duration: optional integer. Maximum clip duration in seconds.
• trim_start_time: optional number. Start time (seconds) for the processing window.
• trim_end_time: optional number. End time (seconds) for the processing window.
• template_id: optional string. ID of a saved template to apply.
• reframing_mode: optional string. One of: fill, fit, automatic (or as configured for the account).
• aspect_ratio: optional string. Target aspect ratio (e.g. 9:16, 1:1; values depend on account configuration).
• include_prompt: optional string. Prompt describing content to include in clips.
• exclude_prompt: optional string. Prompt describing content to exclude from clips.
• transition: optional string. Transition style when combining clips.
• is_extended: optional boolean. When using a video_id with clip indices (e.g. video_id-0), if true, uses extended clip URLs and segments from the parent job.

Response: job_id (string), job_status (e.g. processing).

GET – Get job status and output

Returns the content multiplier job for the given account and video. Response fields include:

• account_id, video_id: strings
• job_id: string (workflow execution ID)
• status: processing | finished | failed
• output: array of clip objects. Each object may include output_url, presigned_url, thumbnail_url, presigned_thumbnail_url, poster_url, presigned_poster_url, extended_clip_url, presigned_extended_url, title, segments_time, virality_score, template_id, liked, discarded, saved, etc.
• error: string (present when status is failed)

While the job is running, status is synced from the workflow; when status is finished, output contains the generated clips. Use the presigned URL fields (presigned_url, presigned_thumbnail_url, presigned_poster_url, presigned_extended_url) to download the corresponding assets; when available, these presigned URLs are the ones that should be used for downloads.

PATCH – Update job output (liked / discarded / saved)

Updates only user-editable fields on each clip in output. Request body:

• output: required array of objects. Length must match the current job’s output length. Each object may contain only liked, discarded, and/or saved (booleans). Other fields are ignored.

Response: the updated job object (same shape as GET).

DELETE – Delete job

Removes the content multiplier job for the account and video. Response: message (e.g. "Content multiplier job deleted successfully").

Get list of AI Content Multiplier jobs

GET https://ingest.api.brightcove.com/v1/accounts/{account_id}/ai/content-multiplier/jobs
Scope: video-cloud/video/read
Response: array of job objects (same shape as the single-job GET)

Get upload URLs for assets

Get presigned POST URLs for uploading assets (e.g. logos, intro/outro videos) used in templates.

GET https://ingest.api.brightcove.com/v1/accounts/{account_id}/ai/content-multiplier/upload-urls/{source_name}
Scope: video-cloud/upload-urls/read

source_name is the file name (or path) to use for the uploaded object. Response: signed_url (POST endpoint), url (S3 URI), form_fields (fields to include in the POST request).

Get / Update Content Multiplier templates

GET https://ingest.api.brightcove.com/v1/accounts/{account_id}/ai/content-multiplier/templates
Scope: video-cloud/video/read

PUT https://ingest.api.brightcove.com/v1/accounts/{account_id}/ai/content-multiplier/templates
Scope: video-cloud/dynamic-ingest/create

GET returns all templates for the account. Each template may include name, logo_url, presigned_logo_url, logo_size, logo_padding, logo_position, intro_asset_url, outro_asset_url, caption settings (captions_enabled, captions_font_name, captions_font_weight, captions_font_size, captions_fill, captions_stroke, captions_stroke_width, captions_background, captions_shadow, captions_color_highlight, captions_animation_type, captions_max_lines, captions_max_chars_per_line, captions_uppercase), aspect_ratio, updated_at.

PUT adds, updates, or removes templates. Request body:

• add: optional object. Map of template name → template object (same fields as above). Adds or replaces templates.
• remove: optional array of strings. Template names to remove.

At least one of add or remove is required. Response: message (e.g. "Templates saved successfully").

CMS Video Details and clips

When using the CMS API (Video Details):

• Clips include a clip_source field with the ID of the source video.
• Videos that have generated clips include a clips field listing the IDs of derived clips.