Motion transfer fails in five specific ways, and four of them are the source clip or a setting rather than the model. A diagnostic order that fixes most attempts without spending more.
3 min read
Listing video is the highest-risk niche for generated content, because a property is a thing a buyer will stand in. The formats that are safe, the hard line on the property itself, and the disclosure question.
2 min read
Per-second API prices for the video models worth using in 2026, what the same clip costs on each, and why credit-based consumer plans are usually the more expensive way to buy the identical model.
4 min read
What platforms currently require when you post AI-generated video, how detection works, and the practical habits creators have settled on. General information, not legal advice.
2 min read
The honest arithmetic: raw model cost for 30 seconds, why nobody generates 30 seconds in one shot, and the retry multiplier that dominates the real number.
3 min read
Toddlers hitting complex choreography with unsettling precision is two models in sequence, not one. The pipeline, why the uncanny precision is the joke, and where it goes wrong.
2 min read
Copying a dance from a Reel onto your own face is a narrow technical job that most 'AI video' tools cannot do at all. The three that can, what they cost per second, and where each one breaks.
4 min read
Four families of AI video model solve four different problems and are all marketed as 'AI video generation'. A decision tree, the inputs each one takes, and what each costs per second.
2 min read
OpenAI retired the Sora app in April 2026 and the API goes dark on 24 September. Here is where the 'me in an AI video' workflow moves, with per-second prices and the real-face policy of each option.
3 min read