“In the end LLMs will continue to exist, no matter what; they will get cheaper. And they will find some utility (indeed they are already have).”
Our reading, the sentence that was checked:
LLMs will continue to exist, get cheaper, and have some utility.
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 69.
- 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-05b853-a1
We recorded 31 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
- 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
- In the end
- LLMs will continue to exist
- no matter what
- they will get cheaper
- they will find some utility
Our interpretation
- Read this as a long-term forecast of persistence, declining comparable-use prices, and at least some practical benefit.
- Cheaper means a lower price for comparable capability and workload, rather than merely a smaller model or a promotional offer.
- Some utility requires a useful deployment or measured benefit; it does not require usefulness on every task.
- The parenthetical acknowledges existing utility but does not supply a deadline for the forecast.
Counts as supported if: Evidence over the intended long-term horizon establishes continuing availability, lower comparable-use prices, and practical utility. Finite observations can support these components without settling indefinite persistence.
Counts as contradicted if: A decisive irreversible event makes continued LLM existence impossible, or establishes that a necessary component cannot occur.
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
Subsequent releases and deployments are consistent with continued existence, useful applications, and cheaper options. They do not settle an indefinite forecast about what happens 'in the end'. No decisive event meets the early-resolution rule.
Limitations
- The deployment and efficiency evidence is vendor-reported.
- A cheaper option is not proof that every model or workload becomes cheaper.
- The forecast has no completion date.
Evidence (2 items)
Context only: Google,
Gemini 3.1 Flash-Lite outperforms 2.5 Flash in speed and quality.
Why it matters here: The release documents continued availability and a claimed improvement in capability and efficiency. A vendor comparison at one date does not establish the full long-term forecast.
- Publisher
- Kind of source
- Company disclosure: the company describing itself, not an independent finding
- Describes events of
- Public from
Context only: Google Cloud,
The core of its text-channel AI agent runs on Flash-Lite.
Why it matters here: Google describes a customer-service deployment and lower costs against comparable thinking-tier models on the same token mix. This is evidence consistent with practical utility and cheaper options, but it is vendor-reported and not a longitudinal market-wide comparison.
- Publisher
- Google Cloud
- Kind of source
- Company disclosure: the company describing itself, not an independent finding
- Describes events of
- Public from
What would settle it
A stated horizon would allow a finite test using matched prices and deployment outcomes. Without one, only a decisive event meeting the early-resolution rule can close the whole forecast.
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
In the end LLMs will continue to exist, no matter what; they will get cheaper. And they will find some utility (indeed they are already have).
- Where
- paragraph 69
- Published
- 2025-11-29
- Source observed
- , live page
- Access
- full text, public
History
Assessment versions
- Assessment marcus-three-years-on-chatgpt-still-isnt-05b853-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.
About assessment marcus-three-years-on-chatgpt-still-isnt-05b853-a1.