“LLMs will always hallucinate, their outputs are not trustworthy as a result, they cannot be deterministic, and any chance of any mistakes of any kind are unforgivable.”
Our reading, the sentence that was checked:
LLMs will always hallucinate, so their outputs are not trustworthy.
Assessment: Unresolved No deadline stated on 2026-10-29 Evidence as of
Where and when Ed Zitron wrote it
Where's Your Ed At, I Will Never Respect A Website, paragraph 6.
- Said
- Source observed
- live page, full text, public
- Evidence as of
- the cutoff for this assessment
- Assessment recorded
- assessment wyea-i-will-never-respect-a-website-d4324f-a1
We recorded 36 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 2026-04-14. 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
- LLMs will always hallucinate
- their outputs are not trustworthy as a result
Our interpretation
- The plain reading is that general-purpose LLMs will retain a propensity to invent false information, not that every answer will be false.
- I read 'not trustworthy' as requiring verification before relying on factual answers. The claim supplies no acceptable error rate or application-specific standard.
- A finite evaluation with zero observed hallucinations does not establish that a model has no propensity to hallucinate.
Counts as supported if: A general technical result establishing unavoidable hallucinations across the claimed class, under assumptions that actually cover its ordinary uses, would support the inevitability assertion. Present-day errors alone cannot establish 'always'.
Counts as contradicted if: A demonstrated general-purpose LLM with a substantiated guarantee against fabricated factual assertions across its claimed operating scope would contradict inevitability. Zero errors on a limited benchmark alone would not suffice.
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 evidence documents hallucination risks, but it cannot establish 'always'. Research also distinguishes theoretical inevitability from practically significant error rates. One study reports zero hallucinations on a small test, which does not prove that the model cannot hallucinate elsewhere. No decisive event meets the early-resolution rule.
Limitations
- The forecast has no deadline.
- Trustworthiness depends on the task and acceptable risk, neither of which Zitron specifies.
- The zero-error result is a preprint reporting a limited historical evaluation.
Evidence (3 items)
Context only: arXiv,
Such "hallucinations" persist even in state-of-the-art systems and undermine trust.
Why it matters here: The paper supports concern about existing systems. Its explanation depends on training and evaluation incentives and does not establish that every future LLM must retain the same problem.
- Publisher
- arXiv
- Kind of source
- Preprint, not peer reviewed
- Describes events of
- Public from
Context only: arXiv,
statistically negligible
Why it matters here: This conditional theoretical result challenges the inference from mathematical inevitability to practical untrustworthiness. It neither demonstrates zero hallucinations in a deployed model nor guarantees that its data assumptions can be met.
- Where
- Abstract, version 3
- Publisher
- arXiv
- Kind of source
- Preprint, not peer reviewed
- Describes events of
- Public from
Context only: arXiv,
zero hallucinations
Why it matters here: The authors report zero hallucinations in an evaluation of 70 academic prompts conducted in February 2025. This is material counterevidence to a claim that hallucinations must occur in every evaluation, but it does not establish elimination of the underlying propensity.
- Publisher
- arXiv
- Kind of source
- Preprint, not peer reviewed
- Describes events of
- Public from
What would settle it
An independently validated general-purpose model with a substantiated guarantee against fabricated factual assertions, or an impossibility result demonstrably covering the full claimed scope.
Where it was said (1)
Said once in the sources this register covers.
: Where's Your Ed At, I Will Never Respect A Website
LLMs will always hallucinate, their outputs are not trustworthy as a result, they cannot be deterministic, and any chance of any mistakes of any kind are unforgivable.
- Where
- paragraph 6
- Published
- 2026-04-14
- Source observed
- , live page
- Access
- full text, public
History
Assessment versions
- Assessment wyea-i-will-never-respect-a-website-d4324f-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 wyea-i-will-never-respect-a-website-d4324f-a1.