“Its products do not scale like traditional software, in that the more users it gets, the more expensive its services are to run because its models are so compute-intensive.”
Our reading, the sentence that was checked:
OpenAI's products do not scale like traditional software, in that the more users it gets, the more expensive its services are to run, because its models are so compute-intensive.
Assessment: Unresolved Evidence as of
Where and when Ed Zitron wrote it
Where's Your Ed At, There Is No AI Revolution, paragraph 8.
- Said
- Source observed
- live page, full text, public
- Evidence as of
- the cutoff for this assessment
- Assessment recorded
- assessment wyea-wheres-the-money-68b867-a1
We recorded 42 claims from this post. See every claim from this post.
The wording above is the version this site observed on 2026-10-28. The source may have been edited after 2025-02-24. This page does not claim the wording was captured then.
What the claim commits to
- Kind
- Causal claim
- Deadline, as the speaker wrote it
- None stated. This site does not add one.
- Deadline state when assessed
- No horizon
- Deadline state on 2026-10-29
- No horizon (the release’s reference date)
From the speaker
- Its products do not scale like traditional software
- the more users it gets, the more expensive its services are to run
- because its models are so compute-intensive
Our interpretation
- Its refers to OpenAI.
- The ordinary reading concerns total serving costs as active usage grows, assuming comparable usage per user and product mix. It does not say that cost per request necessarily rises.
- The comparison implies materially heavier incremental computation than conventional software services, not merely that operating a service has some cost.
- Use Q4 2024, the latest completed calendar quarter before publication, as the reporting proxy for the ongoing operating relationship.
- The claim gives no monetary accounting basis.
Counts as supported if: Matched operating evidence shows that additional active usage materially increases OpenAI's serving costs through inference computation, with a materially heavier incremental cost relationship than comparable conventional software.
Counts as contradicted if: Matched evidence shows that the asserted cost relationship or comparative difference does not hold, or that the observed increase is attributable to another cause rather than additional inference workload.
Retrospective: these criteria were written after the outcome was public.
Current assessment
Does the evidence support the cause?
Outcome: Unresolved
The record was checked, and the evidence available by the cutoff does not settle it.
Scored on a plain reading. The speaker named no measure for this claim, so it was judged on the reading an informed reader would take, stated below under our interpretation. Both reviews had to find that reading fair.
- Evidence as of
- Deadline state when assessed
- No horizon
- Assessment recorded
- Policy version
- 2026-10-28.7
Rationale
The sources support a computational mechanism: advanced model responses require computation. They do not establish the full comparative claim about how OpenAI's operating costs scale with users. Caching also shows that repeated requests can reuse work. Without matched usage, cost and comparator data, the causal assertion remains unresolved.
Limitations
- Compute requirements are not a monetary cost curve.
- Customer prices are not provider costs.
- Efficiency improvements can coexist with increasing total costs.
- The monetary accounting basis is unknown.
Evidence (3 items)
Context only: OpenAI,
significantly more compute
Why it matters here: OpenAI acknowledges substantial computation for advanced capabilities. This supports the proposed mechanism, but does not measure costs as user numbers grow or compare conventional software.
- Where
- Opening paragraph
- Publisher
- OpenAI
- Kind of source
- Company disclosure: the company describing itself, not an independent finding
- Describes events of
- Public from
Context only: OpenAI,
Cached prompts
Why it matters here: The announcement describes reusing previously computed prompt prefixes. This limits any inference that each additional request requires identical new work; customer discounts themselves do not establish provider cost savings.
- Publisher
- OpenAI
- Kind of source
- Company disclosure: the company describing itself, not an independent finding
- Describes events of
- Public from
Context only: Epoch AI,
More transparency
Why it matters here: The analysis models inference energy but explicitly lacks actual data-center measurements. It is later-period context and cannot establish the claimed Q4 operating-cost comparison.
- Publisher
- Epoch AI
- Kind of source
- Independent measurement
- Describes events of
- Public from
What would settle it
A target-period comparison connecting active workload to OpenAI serving costs, controlling for model mix and utilization, alongside similarly measured conventional software and a specified accounting basis.
Where it was said (1)
Said once in the sources this register covers.
: Where's Your Ed At, There Is No AI Revolution
Its products do not scale like traditional software, in that the more users it gets, the more expensive its services are to run because its models are so compute-intensive.
- Where
- paragraph 8
- Published
- 2025-02-24
- Source observed
- , live page
- Access
- full text, public
History
Assessment versions
- Assessment wyea-wheres-the-money-68b867-a1: Current. Outcome: Unresolved. Evidence as of 2026-10-28. Public date not recorded.
Recorded events
Assessment publishedBy this site
First assessment published after both reviews and the final-text check passed.