“It's been eighteen months since this bubble inflated, and since then very little has actually happened involving technology doing new stuff, just an iterative exploration of the very clear limits of what an AI model that generates answers can produce, with the answer being "something that is, at times, sort of good."”
Our reading, the sentence that was checked:
In the 18 months since the AI bubble inflated, generative AI has done very little new, only iterating within clear limits.
Assessment: Unresolved Evidence as of
Where and when Ed Zitron wrote it
Where's Your Ed At, Pop Culture, paragraph 72.
- Said
- Source observed
- live page, full text, public
- Evidence as of
- the cutoff for this assessment
- Assessment recorded
- assessment wyea-pop-culture-d7a1ac-a1
We recorded 27 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 2024-07-08. 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
- eighteen months since this bubble inflated
- very little has actually happened involving technology doing new stuff
- just an iterative exploration
- very clear limits
- something that is, at times, sort of good
Our interpretation
- The ordinary reading is that generative AI had added few substantive capabilities or practical uses during the preceding eighteen months, beyond refinements of existing answer generation.
- Very little permits some advances. A single new application cannot automatically contradict it.
- Use January 8, 2023 through July 8, 2024 by subtracting eighteen months from publication. This operationalizes an approximate interval rather than dating a bubble objectively.
- Count demonstrated or deployed capabilities, while distinguishing them from announced ambitions and distinguishing improved performance from newly enabled tasks.
- Consider both the broad generative-AI wording and the surrounding emphasis on LLMs; do not silently assume that every scientific generative model falls within the intended scope.
Counts as supported if: A representative comparison with the beginning of the interval would show little substantive expansion of reproducible capabilities or practical uses, with developments predominantly improving existing tasks.
Counts as contradicted if: A representative comparison would show substantial expansion into reproducible capabilities or practical uses beyond the baseline, sufficient to challenge the ordinary meaning of very little.
Retrospective: these criteria were written after the outcome was public.
Current assessment
Was it true for the period described?
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 horizon
- Assessment recorded
- Policy version
- 2026-10-28.7
Rationale
The record contains practical deployment, expanded scientific prediction capabilities and serious software-engineering limitations. These findings challenge an absolute claim that nothing new happened, but Zitron said very little. The sources do not establish a representative measure of substantive novelty across the intended field. Treating selected advances as decisive would add a threshold and scope he did not supply; treating a difficult benchmark as proof of stagnation would overstate that evidence.
Limitations
- Company-reported usage does not independently establish effectiveness or novelty.
- AlphaFold 3 may fall outside the intended LLM-centered scope.
- The benchmark result describes the evaluated models and setup, not all systems available at the post date.
- The distinction between substantive novelty and useful iteration remains judgment-dependent.
Evidence (3 items)
Context only: Be My Eyes,
over 5,500
Why it matters here: The company documents use of an application providing visual assistance, making it relevant countercontext to a broad dismissal of practical developments. Usage alone does not establish how novel the tasks were or whether this exceeds very little.
- Where
- Opening paragraph
- Publisher
- Be My Eyes
- Kind of source
- Company disclosure: the company describing itself, not an independent finding
- Describes events of
- Public from
Context only: Nature,
across biomolecular space
Why it matters here: The paper describes a diffusion-based model covering joint structures involving proteins, nucleic acids and other molecules, with improvements over specialized tools. This challenges a broad dismissal of generative-model developments, but its relevance depends on whether scientific models fall within the intended LLM-centered scope.
- Publisher
- Nature
- Kind of source
- Peer-reviewed paper
- Describes events of
- Public from
Context only: arXiv,
a mere $1.96$%
Why it matters here: The reported SWE-bench evaluation documents substantial difficulty with real repository issues. It supports a narrower concern about model limitations, but does not measure industry-wide novelty or establish performance for every system available by July 2024.
- Publisher
- arXiv
- Kind of source
- Preprint, not peer reviewed
- Describes events of
- Public from
What would settle it
A representative comparison of capabilities and deployed uses at the beginning and end of the interval, with reproducible evidence of newly enabled tasks. A decisive judgment also requires a defensible interpretation of very little and the intended model scope.
Where it was said (1)
Said once in the sources this register covers.
: Where's Your Ed At, Pop Culture
It's been eighteen months since this bubble inflated, and since then very little has actually happened involving technology doing new stuff, just an iterative exploration of the very clear limits of what an AI model that generates answers can produce, with the answer being "something that is, at times, sort of good."
- Where
- paragraph 72
- Published
- 2024-07-08
- Source observed
- , live page
- Access
- full text, public
History
Assessment versions
- Assessment wyea-pop-culture-d7a1ac-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.