“And if Sustekver and I (and at this point many others) are right that scaling is largely played out, a decline in demand is sure to follow.”
Our reading, the sentence that was checked:
If scaling is largely played out, a decline in demand for AI compute is sure to follow.
Assessment: Unresolved No deadline stated on 2026-10-29 Evidence as of
Where and when Gary Marcus wrote it
Marcus on AI, Three years on, ChatGPT still isn't what it was cracked up to be – and it probably never will be, paragraph 56.
- Said
- Source observed
- live page, full text, public
- Evidence as of
- the cutoff for this assessment
- Assessment recorded
- assessment marcus-three-years-on-chatgpt-still-isnt-1ad55b-a2
We recorded 30 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 2025-11-29. This page does not claim the wording was captured then.
What the claim commits to
- Kind
- Conditional 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)
Conditions
- if scaling is largely played out (antecedent, stated by the speaker)
From the speaker
- if ... scaling is largely played out
- a decline in demand
- is sure to follow
Our interpretation
- The surrounding discussion makes AI compute demand the target.
- I read the condition as substantially exhausted capability gains from additional scale, rather than merely diminishing returns.
- I read demand as aggregate demand for AI computation, rather than business adoption rates or one supplier's revenue.
- The claim supplies neither a deadline nor a lag between the condition and the predicted decline.
Counts as supported if: Independent evidence establishes that scaling has largely ceased delivering substantial capability gains, followed by a sustained decline in aggregate AI compute demand.
Counts as contradicted if: Evidence establishes the antecedent and decisively refutes the asserted necessary decline. A finite period of continued demand growth cannot alone contradict this forecast because no deadline was given.
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
The April 9, 2026 analysis describes diminishing returns and the efficiency improvements needed for continued scaling; it does not establish that scaling has largely stopped delivering substantial capability gains [e1]. It also provides no aggregate demand series showing a sustained decline after such a plateau [e1]. The evidence therefore leaves the conditional forecast unresolved [e1].
Limitations
- The evidence does not establish the scaling condition or measure subsequent aggregate AI compute demand.
- The claim specifies no threshold for exhausted scaling gains, deadline, or response lag.
- A finite period of continued demand growth cannot settle this undated forecast against the speaker.
Evidence (1 item)
Context only: arXiv,
diminishing returns mean rising operational burden
Why it matters here: This theoretical analysis discusses the efficiency improvements needed for continued scaling. It does not establish that additional scale has largely stopped producing substantial capability gains.
- Where
- Abstract, version 2
- Publisher
- arXiv
- Kind of source
- Preprint, not peer reviewed
- Describes events of
- Public from
What would settle it
Independent measurements establishing broadly exhausted scaling gains, followed by comparable aggregate physical compute-demand measurements showing a sustained decline, would support the forecast. Contradiction would require establishing the condition and decisively refuting the necessary decline; finite continued growth alone would not suffice.
Where it was said (1)
Said once in the sources this register covers.
: Marcus on AI, Three years on, ChatGPT still isn't what it was cracked up to be – and it probably never will be
And if Sustekver and I (and at this point many others) are right that scaling is largely played out, a decline in demand is sure to follow.
- Where
- paragraph 56
- Published
- 2025-11-29
- Source observed
- , live page
- Access
- full text, public
History
Assessment versions
- Assessment marcus-three-years-on-chatgpt-still-isnt-1ad55b-a2: 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.
About assessment marcus-three-years-on-chatgpt-still-isnt-1ad55b-a2.