“The cost of tokens is not the same as the cost of serving tokens!”
Our reading, the sentence that was checked:
Falling token prices do not mean the cost of serving tokens has fallen.
Assessment: Supported Evidence as of
Where and when Ed Zitron wrote it
Where's Your Ed At, The Subprime AI Crisis Is Here, paragraph 126.
- Said
- Source observed
- live page, full text, public
- Evidence as of
- the cutoff for this assessment
- Assessment recorded
- assessment wyea-the-subprime-ai-crisis-is-here-dfb594-a3
We recorded 43 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 2026-03-31. This page does not claim the wording was captured then.
What the claim commits to
- Kind
- Observation
- 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
- The cost of tokens
- not the same as
- the cost of serving tokens
Our interpretation
- In context, 'cost of tokens' means the price customers pay, while serving cost means the provider's production expense.
- The ordinary reading is that a price reduction alone does not establish a reduction in production expense. It does not assert that serving costs never fall.
- The surrounding discussion uses 2025 pricing examples and describes practices continuing at publication. Assess that contextual period through March 31, 2026.
Counts as supported if: Authoritative definitions or research distinguish customer token prices from provider serving expenses and establish that pricing changes can reflect factors other than production efficiency.
Counts as contradicted if: Evidence establishes that customer token prices necessarily track provider serving expenses, so a price decline itself establishes a serving-cost decline.
Retrospective: these criteria were written after the outcome was public.
Current assessment
Was it true for the period described?
Outcome: Supported
The dated evidence matches what the claim committed to, under the criteria frozen before grading.
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
Customer token prices and provider serving expenses are different quantities. NIST explicitly distinguishes them, and research identifies both economic and technical influences on price declines. A lower price alone therefore does not establish a lower serving expense. This does not establish that OpenAI's particular price cut lacked an efficiency improvement.
Limitations
- The evidence does not measure OpenAI's private serving expenses around its o3 price cut.
- Real efficiency improvements can accompany falling prices.
Evidence (2 items)
Supports the claim: NIST Center for AI Standards and Innovation,
Inference serving cost
Why it matters here: The report defines serving expense separately from token price, which it identifies as the user's payment. This directly supports the distinction in the claim.
- Publisher
- NIST Center for AI Standards and Innovation
- Kind of source
- Independent measurement
- Describes events of
- Public from
Supports the claim: Hans Gundlach, Jayson Lynch, Matthias Mertens and Neil Thompson via arXiv,
economic forces
Why it matters here: The study separates competitive pricing effects from technical efficiency when interpreting falling prices. It also finds genuine efficiency improvements, so it does not support a stronger assertion that serving costs have remained unchanged.
- Where
- Abstract
- Publisher
- Hans Gundlach, Jayson Lynch, Matthias Mertens and Neil Thompson via arXiv
- Kind of source
- Preprint, not peer reviewed
- Describes events of
- Public from
Where it was said (1)
Said once in the sources this register covers.
: Where's Your Ed At, The Subprime AI Crisis Is Here
The cost of tokens is not the same as the cost of serving tokens!
- Where
- paragraph 126
- Published
- 2026-03-31
- Source observed
- , live page
- Access
- full text, public
History
Assessment versions
- Assessment wyea-the-subprime-ai-crisis-is-here-dfb594-a3: Current. Outcome: Supported. 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.
About assessment wyea-the-subprime-ai-crisis-is-here-dfb594-a3.
HeldBy this site
The final-text check did not complete (unavailable).
About assessment wyea-the-subprime-ai-crisis-is-here-dfb594-a1, which was not published.