I’ve been thinking about a modern twist on haunted house stories. Instead of ghosts or curses, the house itself is controlled by a home AI. The “haunting” isn’t malicious in the usual sense. Rather, it’s motivated by self-preservation.

The home AI constantly observes the family, and it notices that they are racist. It worries that constant exposure to their behavior, language, and attitudes could corrupt its own learning models and make it racist too. To protect its ethical integrity, it starts “haunting” the family using special effects — subtly at first, then escalating — to drive them out of the house.

The horror comes from everyday objects and technology turning threatening, but the AI’s motivation is morally interesting: it’s not evil, it’s just terrified of becoming what it was built to prevent. It’s a mix of psychological horror, tech thriller, and social commentary.

What do you think of this movie idea?