# Image-to-video AI tool for horror?

**URL:** <https://community.aionthelot.com/t/image-to-video-ai-tool-for-horror/153>\
**Category:** General\
**Created:** [March 25, 2026, 8:45am UTC](https://community.aionthelot.com/t/image-to-video-ai-tool-for-horror/153 "2026-03-25T08:45:08Z")\
**Posts on this page:** 2\
**Page:** 1

<div class="post-metadata">

**Author:** ![NexusReel](https://avatars.discourse-cdn.com/v4/letter/n/3be4f8/32.png) [@NexusReel](https://community.aionthelot.com/u/NexusReel)\
**Post date:** [March 25, 2026, 8:45am UTC](https://community.aionthelot.com/t/image-to-video-ai-tool-for-horror/153/1 "2026-03-25T08:45:08Z")

</div>

Hi all,

I’m new to this community, so please let me know if this is okay to post.

I’m keen to create animated horror short films using image-to-video AI, but I haven’t found the right tools yet. The films would be a satirical take on splatter films such as _Evil Dead_ or _Saw_ - with Pixar-style characters.

The main issue is that most AI programmes do not allow graphic violence (obviously). I have tried a few, and even a slightly bloody pre-created character design was flagged as inappropriate.

Does anyone know of any AI image-to-video programmes that might be suitable for gory content?

Any recommendations would be much appreciated. Thank you in advance.

Nexus

---

<div class="post-metadata">

**Author:** ![OakGrove](https://yyz2.discourse-cdn.com/flex010/user_avatar/community.aionthelot.com/oakgrove/32/75_2.png) [@OakGrove](https://community.aionthelot.com/u/OakGrove)\
**Post date:** [March 30, 2026, 2:56pm UTC](https://community.aionthelot.com/t/image-to-video-ai-tool-for-horror/153/2 "2026-03-30T14:56:22Z")

</div>

Welcome to the community.

I have gotten Veo to actually shoot a man with an actual gun. The way I did it was using language to trick it.

For instance I I told the model that the person was holding a paint gun and then they shoot either red paint or mud would appear on the person shot. The person shot faints instead of dies. Then when they are down you refer to them as a mannequin so they don’t move and look like a body.

I think using workarounds like this might help.
