I kept running into the same thing with AI video: you write a prompt, you get a clip, and then it's over. One ending, no way in.
Alternate Realities is the opposite. Start from a prompt, photo, or short video. Point at something in the frame — or write what happens next — and it continues as real video from exactly there. The final frame becomes the next scene, which keeps the world visually continuous.
Everything is private by default. If you make a world public, anyone can walk your path or take a different one.
Honest notes:
Try a public world without signing in:
https://alternaterealities.lol/world/pip-and-the-star-letter-a2b21418
Site: https://alternaterealities.lol
Also launching on Product Hunt tomorrow (Tue): https://www.producthunt.com/products/alternate-realities?launch=alternate-realities
Would love feedback on whether the branching mechanic is clear in the first 30 seconds.
The branching mechanic is the interesting part, but the real signal seems to be whether people create multiple branches rather than just watch one path. Have early users actually explored several branches in the same world, or is that behavior still what you're trying to validate?