“Altman had, much to my surprise, just echoed my longstanding position that current techniques alone would not be enough to get to AGI”
Our reading, the sentence that was checked:
Current AI techniques alone will not be enough to reach AGI.
Assessment: Unresolved No deadline stated on 2026-10-29 Evidence as of
Where and when Gary Marcus wrote it
Marcus on AI, Gary Marcus vs Sam Altman on GPT-o1 aka Strawberry, paragraph 4.
- Said
- Source observed
- live page, full text, public
- Evidence as of
- the cutoff for this assessment
- Assessment recorded
- assessment marcus-gary-marcus-vs-sam-altman-on-gpt-3723f0-a2
We recorded 5 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 2024-09-12. 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)
Conditions
- current techniques alone (qualifier, stated by the speaker)
From the speaker
- current techniques alone
- would not be enough
- get to AGI
Our interpretation
- Assess the underlying forecast, not the separate assertion that Altman agreed with Marcus.
- Read current techniques as the LLM training and scaling approaches under discussion in 2023–24.
- Read AGI as broadly human-level general intellectual capability, including adaptation to unfamiliar tasks.
- The restriction to current techniques defines the systems covered; it is not a future event that must trigger the forecast.
Counts as supported if: A broadly accepted result establishes that the specified techniques cannot achieve human-level general intelligence without additional methods.
Counts as contradicted if: An independently verified AGI system achieves that capability using only the specified techniques.
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
ARC Prize's September 3, 2026 report documents strong Astra performance on ARC-AGI-3, but does not establish AGI or exclusive use of the techniques covered by this claim [e2]. The September 2026 AGI Society assessment also finds that the published evidence does not establish Astra as AGI [e1]. Neither item establishes that the specified techniques cannot achieve AGI, so this evidence settles neither side of the forecast [e1, e2].
Limitations
- There is no generally accepted empirical test for AGI [e1].
- The benchmark report does not establish the training-method lineage required to test the claim [e2].
- These assessments do not establish a general impossibility result [e1, e2].
Evidence (2 items)
Context only: AGI Society,
no generally accepted empirical test
Why it matters here: The assessment considers Huang's and Brockman's AGI declarations but concludes that the published evidence does not yet establish GPT-6 Astra as AGI. This documents disagreement rather than proving Marcus's forecast.
- Publisher
- AGI Society
- Kind of source
- Secondary summary
- Describes events of
- Public from
Context only: ARC Prize,
96% of levels.
Why it matters here: Astra's independently measured progress is material evidence against assuming that these approaches have stopped improving. ARC Prize nevertheless says its bounded benchmark does not establish AGI, and this report does not establish the required training-method lineage.
- Publisher
- ARC Prize
- Kind of source
- Independent measurement
- Describes events of
- Public from
What would settle it
Independent verification of broadly human-level general intelligence achieved using only the specified 2023–24 techniques, supported by detailed technical disclosures; or a broadly accepted general result establishing that those techniques cannot suffice without additional methods.
Where it was said (1)
Said once in the sources this register covers.
: Marcus on AI, Gary Marcus vs Sam Altman on GPT-o1 aka Strawberry
Altman had, much to my surprise, just echoed my longstanding position that current techniques alone would not be enough to get to AGI
- Where
- paragraph 4
- Published
- 2024-09-12
- Source observed
- , live page
- Access
- full text, public
History
Assessment versions
- Assessment marcus-gary-marcus-vs-sam-altman-on-gpt-3723f0-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-gary-marcus-vs-sam-altman-on-gpt-3723f0-a2.