“Even the latest systems like Deep Research are still struggling in a few ways – and those ways pretty much correspond exactly to the places that I warned would be LLM’s Achilles’ Heels: hallucinations and reasoning errors.”
Our reading, the sentence that was checked:
The latest systems such as OpenAI's Deep Research still suffer from hallucinations and reasoning errors.
Assessment: Supported Evidence as of
Where and when Gary Marcus wrote it
Marcus on AI, Five ways in which the last 3 months — and especially the DeepSeek era — have vindicated “Deep learning is hitting a wall", paragraph 7.
- Said
- Source observed
- live page, full text, public
- Evidence as of
- the cutoff for this assessment
- Assessment recorded
- assessment marcus-five-ways-in-which-the-last-3-months-471130-a1
We recorded 8 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 2025-02-09. 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
- Even the latest systems like Deep Research
- still struggling
- hallucinations and reasoning errors
Our interpretation
- The plain reading is that the launch version of OpenAI Deep Research still produced fabricated factual material and incorrect inferences.
- The claim does not specify an error rate or say that the system was useless or worse than its predecessors.
- The target period is its launch week through the post date, the latest public observation period available when Marcus wrote.
- This assessment concerns the stated remaining limitations, not Marcus's separate claim to prior foresight.
Counts as supported if: Documented launch-period outputs, evaluations, or a direct developer disclosure establish that both hallucinations and incorrect inferences remained possible.
Counts as contradicted if: Reliable evidence establishes that the relevant system did not exhibit either alleged error class and that purported examples were misattributed or invalid.
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
OpenAI's own launch disclosure acknowledges that Deep Research could hallucinate facts and make incorrect inferences. That directly supports the ordinary reading that these problems remained. The disclosure also says their rate had fallen, so this finding does not imply that the system had made no progress.
Limitations
- Supports the existence of both error classes, not any particular frequency or severity.
- Does not independently validate the tariff-paper or temporal-reasoning examples.
- Does not establish Marcus's claim of predictive priority.
Evidence (1 item)
Supports the claim: OpenAI,
but it’s still early and has limitations.
Why it matters here: The paragraph expressly identifies factual hallucinations and incorrect inferences as remaining limitations. It also reports a lower rate than existing ChatGPT models, which qualifies the severity without contradicting their continued presence.
- Publisher
- OpenAI
- Kind of source
- Company disclosure: the company describing itself, not an independent finding
- Describes events of
- Public from
Where it was said (1)
Said once in the sources this register covers.
: Marcus on AI, Five ways in which the last 3 months — and especially the DeepSeek era — have vindicated “Deep learning is hitting a wall"
Even the latest systems like Deep Research are still struggling in a few ways – and those ways pretty much correspond exactly to the places that I warned would be LLM’s Achilles’ Heels: hallucinations and reasoning errors.
- Where
- paragraph 7
- Published
- 2025-02-09
- Source observed
- , live page
- Access
- full text, public
History
Assessment versions
- Assessment marcus-five-ways-in-which-the-last-3-months-471130-a1: 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 marcus-five-ways-in-which-the-last-3-months-471130-a1.