# Video stops being write-only

Period: Feb – Apr 2025 · Era: 2025 · The Studio Era

Swap, add, keyframe and perform become verbs.

Video stopped being something you could only generate from scratch. New tools let you swap a person in an existing clip, add an object, transition between two frames, or move a real actor's performance onto a generated character.

## How it worked

1. Pika 2.2: insert a subject into existing footage (Pikadditions), replace one (Pikaswaps), or keyframe a transition up to ten seconds (Pikaframes).
2. Talking characters from a single photo plus audio: OmniHuman-1, Hedra Character-3.
3. Camera as a menu: Higgsfield's fifty-plus motion presets.
4. Runway Gen-4 References: up to three reference images (photos, generated images, 3D models or selfies) for consistent characters and locations, then image-to-video.
5. The showrunner pipeline: webcam performance → Act-One → Hedra lip-sync → ElevenLabs voices → Premiere, about twelve hours per episode.

## Why it mattered

Reference-conditioned consistency plus swap and insert tools turned generation into editing.

## Tools of the era

- Pika 2.2 (Pikadditions, Pikaswaps, Pikaframes) (2025): https://pika.art
- Hedra Character-3 (2025): https://www.hedra.com
- Higgsfield (2025): https://higgsfield.ai
- Runway Gen-4 (2025): https://runway.com/research/introducing-runway-gen-4
- Gen-4 References (2025): https://runway.com

## Example videos and posts

- Gen-4 References · Runway: https://x.com/runwayml/status/1917628723903463526
- How Neural Viz makes an entire universe · Neural Viz: https://x.com/NeuralViz/status/1921732202934788551
- Runway and Midjourney references: tips and showdown · Theoretically Media: https://www.youtube.com/watch?v=Poy__YfsQNo
- Create cinematic AI videos with Higgsfield · Curious Refuge: https://www.youtube.com/watch?v=Lm5L2Atw3S0

## On the timeline

- [Runway Gen-4](https://aifilmhistory.org/entry/runway-gen4) (2025-03)

## Sources

- Runway Gen-4 launch: https://techcrunch.com/2025/03/31/runway-releases-an-impressive-new-video-generating-ai-model
- Hedra Character-3: https://www.hedra.com/models/video/hedra/character-3
- Higgsfield camera controls: https://higgsfield.ai/camera-controls
- The showrunner (Neural Viz): https://www.thedaringcreatives.com/creator-stories/the-showrunner-how-neural-viz-makes-an-entire-tv-universe-with-ai/

---
Source: https://aifilmhistory.org/workflows/editing-primitives