“Hallucinations are a huge, unsolved issue.”
Our reading, the sentence that was checked:
LLM hallucinations are a huge, unsolved problem.
Assessment: Supported Evidence as of
Where and when Gary Marcus wrote it
Marcus on AI, Top 5 reasons OpenAI is probably not worth 90 billion dollars, paragraph 9.
- Said
- Source observed
- live page, full text, public
- Evidence as of
- the cutoff for this assessment
- Assessment recorded
- assessment marcus-top-5-reasons-openai-is-probably-5ff0ce-a1
We recorded 9 claims from this post. See every claim from this post.
The wording above is the version this site observed on 2026-10-29. The source may have been edited after 2023-09-27. 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
- Hallucinations
- a huge, unsolved issue
Our interpretation
- The plain reading is that fabricated or incorrect factual output remained a substantial practical reliability problem for contemporary general-purpose LLMs.
- Huge means consequential enough to require care and safeguards in important applications; Marcus supplies no numerical error-rate threshold.
- Unsolved means that available models and mitigations had not eliminated the general problem. It does not mean that no improvements or useful applications existed.
- The target is the latest public 2023 model-reliability picture through the post date. Model evaluations have no fixed quarterly reporting period.
Counts as supported if: Contemporaneous primary evaluations or model disclosures identify continuing hallucinations and practical limitations serious enough to require safeguards in consequential uses.
Counts as contradicted if: Robust contemporaneous evidence establishes that hallucinations had been reliably eliminated across the relevant general-purpose uses, or reduced to a negligible practical issue.
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 contemporary GPT-4 disclosure explicitly acknowledges hallucinations and consequential reliability limits. Its recommendations for human review, additional grounding, or avoiding some high-stakes uses support the ordinary reading of a substantial unresolved problem. Reported improvements do not contradict that reading.
Limitations
- Huge is a qualitative judgment; no universal hallucination rate is established.
- This assesses the 2023 state, not whether hallucinations remain equally serious in later systems.
- The evidence supports persistence of the problem, not impossibility of solving it.
Evidence (1 item)
Supports the claim: OpenAI,
it “hallucinates” facts
Why it matters here: OpenAI identifies continuing factual unreliability and recommends safeguards or avoiding some high-stakes uses. The same section reports substantial improvement over preceding models, which limits any reading that hallucination mitigation had made no progress.
- 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, Top 5 reasons OpenAI is probably not worth 90 billion dollars
Hallucinations are a huge, unsolved issue.
- Where
- paragraph 9
- Published
- 2023-09-27
- Source observed
- , live page
- Access
- full text, public
History
Assessment versions
- Assessment marcus-top-5-reasons-openai-is-probably-5ff0ce-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-top-5-reasons-openai-is-probably-5ff0ce-a1.