Strategy question

Sora's standalone app was discontinued in 2026. How would you decide whether to relaunch video generation as a standalone product vs. a ChatGPT feature?

Practice this question out loud. An AI interviewer asks it, follows up like a real interviewer would, and scores your answer. Type or speak.

What this question tests

Strategic judgment on product surface area: standalone app versus a feature nested in the flagship product.

State the core question: does video generation need its own habit loop and discovery surface, or does it work better inside ChatGPT where distribution is already massive.
Review why it failed standalone: a separate app means starting user acquisition from zero and asking users to form a second habit.
Weigh the case for a feature: ChatGPT's existing hundreds of millions of users get instant access, and video becomes one more reason to stay in one app.
Weigh the case for standalone: video generation has a distinct, more visual, feed-like usage pattern that a chat interface may constrain.
Recommend embedding as a ChatGPT feature first to leverage distribution, with a possible standalone spinoff later only if usage data shows a genuinely different behavior pattern.

Diagnoses the standalone failure correctly as a distribution and habit-formation problem, not necessarily a product-quality problem.
Recommends re-launching inside ChatGPT to inherit its existing engagement and reduce acquisition cost to near zero.
Proposes a concrete test: track whether video generation usage inside ChatGPT clusters into a distinct behavior pattern, like session length or repeat same-day use, before considering standalone again.
Names the tradeoff: a chat interface may be a poor fit for a visual, iterative creative workflow, and flags that as a risk to monitor.

Recommending standalone relaunch without addressing why distribution failed the first time.
Ignoring that a feature nested in chat may cap the product's ceiling if video needs a different interaction model.

What early signal would tell you video generation deserves to spin out again?
How would you design the ChatGPT UI differently for a visual generation feature versus text chat?