“While Sora and other video generators like Pika may seem like the future (and are capable of some impressive magic tricks), they are not particularly adept — much like a lot of generative AI — at performing a particular task.”
Our reading:
Sora, Pika and much generative AI are not particularly adept at performing a particular task.
Assessment: Supported
What happened
Was it true for the period described?
Outcome: Supported
The evidence matches what they said.
Judged on a plain reading. They named no measure, so we used the reading a typical informed reader would take. It is spelled out under “How we judged it”.
Evidence up to . Checked .
Why
Read with the sentences that follow it, the claim is that Sora and Pika could not reliably produce a specific requested video with consistent details. OpenAI's Sora launch page of February 15, 2024 (e1) says the model 'may also confuse spatial details included in a prompt, such as discerning left from right, or struggle with precise descriptions of events that unfold over time, like specific camera trajectories.' A hands-on review of Pika 1.0 published on December 22, 2023 (e2) found 'no guarantee' that a follow-on clip would 'look anything like the first.' It reported similar problems getting Pika to follow specific style suggestions and said users must deal with 'a lot of trial and error.' The same OpenAI page also says Sora maintains 'adherence to the user's prompt' (e3). That is the developer's own description, not an evaluation, and the page also lists prompt-following limits. On the plain reading, this evidence supports the claim.
Caveats
- The phrase 'not particularly adept' is vague and sets no threshold. A different reading could change the result.
- Sora was not public during the period. The only Sora evidence is the developer's own disclosure, and no independent tests exist.
- The Pika evidence comes from one hands-on opinion review. A more positive Tom's Guide review was not read.
- No task-based benchmark of either tool from the period was found.
Evidence (3 items)
Supports the claim: OpenAI,
It may struggle to simulate the physics of a complex scene, and may not comprehend specific instances of cause and effect (for example: a cookie might not show a mark after a character bites it). The model may also confuse spatial details included in a prompt, such as discerning left from right, or struggle with precise descriptions of events that unfold over time
Why it matters: The developer admits Sora may confuse spatial details in a prompt and struggle with precise events and camera paths. In other words, it may fail at a specific requested task.
- Publisher
- OpenAI
- Kind of source
- Company disclosure: the company describing itself, not an independent finding
- About events on
- Published
Supports the claim: Inverse,
This is an issue with all current generative AI video tools, but if you want continuity between clips on Pika, you’re out of luck.
Why it matters: A hands-on reviewer found Pika 1.0 could not keep continuity between clips or follow style requests reliably. This supports the claim for Pika.
- Publisher
- Inverse
- Kind of source
- News report
- About events on
- Published
Context only: OpenAI,
Sora can generate videos up to a minute long while maintaining visual quality and adherence to the user’s prompt.
Why it matters: OpenAI's marketing claim of prompt adherence cuts the other way. It is a self-description, not an evaluation, and the same page lists prompt-following failures.
- Publisher
- OpenAI
- Kind of source
- Company disclosure: the company describing itself, not an independent finding
- About events on
- Published
How we judged it
- Kind of claim
- Observation
- Deadline
- None given. We don’t add one.
Where we had to interpret
- Plain reading, using the next sentences ('almost-useful video', consistency 'from clip to clip'): video generators like Sora and Pika could not reliably produce a specific, requested result, such as following a detailed prompt or keeping continuity, without errors.
- Target period: the state of these tools at the post date (November 2023, when Pika 1.0 launched, through February 19, 2024, after Sora was announced). Sora was not public then, so developer disclosures and hands-on reviews of Pika are the available evidence.
- The 'much like a lot of generative AI' aside is not scored separately.
Right if: Developer disclosures or hands-on evaluations from the period showing Sora or Pika often failed to follow specific instructions or keep consistency, making them unreliable for a given task.
Wrong if: Evaluations from the period showing Sora or Pika reliably produced the specific requested output with consistency suitable for practical use.
We wrote this test after the outcome was known.
We found 31 claims in this post. See them all.
Full record
Source and dates
Where's Your Ed At, Subprime Intelligence, paragraph 11.
- Said
- We read the source
- live page, full text, public
- Evidence up to
- Checked
We read this wording on 2026-10-28. The post may have been edited after 2024-02-19.
Where it was said (1)
Said once in the posts we read.
: Where's Your Ed At, Subprime Intelligence
While Sora and other video generators like Pika may seem like the future (and are capable of some impressive magic tricks), they are not particularly adept — much like a lot of generative AI — at performing a particular task.
- Where
- paragraph 11
- Published
- 2024-02-19
- We read it
- , live page
- Access
- full text, public
History
Versions
- Current: Supported. Evidence up to 2026-10-28.
What happened to this record
PublishedBy us
Published after both AI reviews and the final check passed.