“Moreover, the scale is different, and serving a webpage will always cost less than processing a request sent to a generative AI model.”
Our reading, the sentence that was checked:
Processing a generative AI request will always cost more than serving a webpage.
Assessment: Unresolved No deadline stated on 2026-10-29 Evidence as of
Where and when Ed Zitron wrote it
Where's Your Ed At, Subprime Intelligence, paragraph 30.
- Said
- Source observed
- live page, full text, public
- Evidence as of
- the cutoff for this assessment
- Assessment recorded
- assessment wyea-sam-altman-fried-dc9d67-a1
We recorded 26 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 2024-02-19. This page does not claim the wording was captured then.
What the claim commits to
- Kind
- Forecast
- Deadline, as the speaker wrote it
- None stated. This site does not add one.
- Deadline state when assessed
- No deadline stated
- Deadline state on 2026-10-29
- No deadline stated (the release’s reference date)
From the speaker
- serving a webpage
- will always cost less
- processing a request sent to a generative AI model
Our interpretation
- The surrounding discussion concerns infrastructure providers and application operators. The ordinary reading compares their serving costs, not users' mobile-data charges.
- Always asserts a permanent cost ordering, not merely a typical difference in February 2024.
- A fair comparison must specify the webpage, model request, utilization and cost boundary. Neither free promotional pricing nor an unusually wasteful webpage is sufficient by itself.
Counts as supported if: Comparable measurements may support the ordering for measured workloads and dates, but finite observations cannot establish that it will always hold.
Counts as contradicted if: A reproducible comparison demonstrates a genuine generative-model request costing no more to serve than an ordinary webpage under a consistent provider-side cost boundary.
Retrospective: these criteria were written after the outcome was public.
Current assessment
Did the forecast come true?
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 deadline stated
- Assessment recorded
- Policy version
- 2026-10-28.7
Rationale
No decisive counterexample meeting the stated rule was established. The Sierra Leone study finds AI access cheaper than loading webpages, but includes users' mobile-data charges and AI API prices. Treating that as a measurement of providers' serving costs would change the comparison. The forecast therefore remains open; elapsed time without a qualifying counterexample does not prove always.
Limitations
- A broader user-price reading would make the study substantially more adverse to the claim.
- Unspecified workloads and cost allocation limit comparisons.
Evidence (1 item)
Context only: Daniel Björkegren and coauthors / arXiv,
8 times more cost effective
Why it matters here: The study compares mobile-data charges for loading webpages with AI token prices plus mobile-data charges. It challenges a user-cost reading, but does not measure both providers' underlying serving expenses.
- Publisher
- Daniel Björkegren and coauthors / arXiv
- Kind of source
- Preprint, not peer reviewed
- Describes events of
- Public from
What would settle it
A dated, reproducible comparison showing generative inference at or below ordinary webpage-serving cost under a consistent provider-side cost boundary would contradict the claim.
Where it was said (1)
Said once in the sources this register covers.
: Where's Your Ed At, Subprime Intelligence
Moreover, the scale is different, and serving a webpage will always cost less than processing a request sent to a generative AI model.
- Where
- paragraph 30
- Published
- 2024-02-19
- Source observed
- , live page
- Access
- full text, public
History
Assessment versions
- Assessment wyea-sam-altman-fried-dc9d67-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.