“On the other hand, the use cases on the negative side of the ledger are proliferating fast, literally faster than I can report them here, from fake books (undermining authors) to rings of fake websites designed to sell ads”
Our reading, the sentence that was checked:
Harmful uses of generative AI (fake books, ad-farm websites, election misinformation, polluted science) are proliferating rapidly.
Assessment: Supported Evidence as of
Where and when Gary Marcus wrote it
Marcus on AI, The race between positive and negative applications of Generative AI is on – and not looking pretty, paragraph 5.
- Said
- Source observed
- live page, full text, public
- Evidence as of
- the cutoff for this assessment
- Assessment recorded
- assessment marcus-the-race-between-positive-and-negative-0c70cf-a1
We recorded 8 claims from this post. See every claim from this post.
The wording above is the version this site observed on 2026-10-29. The source may have been edited after 2024-03-29. 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
- use cases on the negative side of the ledger
- proliferating fast
- fake books
- rings of fake websites designed to sell ads
- election misinformation
- articles partially written by GPT in which hallucinations may filter in
Our interpretation
- The ordinary reading is rapid expansion of harmful applications across several domains, not an identical growth rate in every listed category.
- The comparison with Marcus's personal reporting capacity is rhetorical and has no supplied measurable baseline.
- The target is the current quarter through publication, the latest interval for which incident reporting could be available. Earlier counts supply comparison baselines.
- The science clause identifies a possible route for errors; it does not assert that every AI-assisted paper contains hallucinations.
Counts as supported if: Dated counts show substantial recent expansion in harmful applications, with contemporaneous documentation of their spread across the named domains.
Counts as contradicted if: Comparable evidence shows that the asserted expansion did not occur, or that the named applications were materially misidentified.
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
The ordinary qualitative reading is supported. Identified unreliable AI sites expanded sharply, reporting described growing book imitations, and an official investigation documented election misinformation using apparent AI voice cloning. Research also found increasing LLM-assisted scientific writing, although it did not establish widespread hallucinations. This supports spreading harmful applications, not a measured common growth rate or Marcus's personal reporting comparison.
Limitations
- Support depends on reading proliferation as broad expansion rather than proven rapid growth in every category.
- News reporting supplies the dated website count and book cases.
- The official-release reproduction is classified as a disclosure because the schema has no government-disclosure category.
- AI use in science is not itself proof of harm.
Evidence (4 items)
Supports the claim: The Guardian Nigeria,
false claims
Why it matters here: The report records 713 identified AI-generated sites, compared with fewer than 50 in May and around 600 in December 2023, and describes advertising incentives. These are discovery counts, not a controlled census of site creation.
- Publisher
- The Guardian Nigeria
- Kind of source
- News report
- Describes events of
- Public from
Supports the claim: WIRED,
spammy summary industry.
Why it matters here: Reporting documents multiple apparent AI imitations and summaries, with publishing specialists describing rapid growth. AI attribution partly relies on detection software and informed judgments.
- Where
- Summary Execution
- Publisher
- WIRED
- Kind of source
- News report
- Describes events of
- Public from
Supports the claim: New Hampshire Office of the Attorney General; reproduced by Democracy in Action,
illegal robocalls
Why it matters here: The official release describes apparent AI voice cloning used to discourage voting. It establishes an election application, but does not independently measure its growth rate.
- Publisher
- New Hampshire Office of the Attorney General; reproduced by Democracy in Action
- Kind of source
- Company disclosure: the company describing itself, not an independent finding
- Describes events of
- Public from
Context only: Andrew Gray / arXiv,
rapidly increased
Why it matters here: The study finds growing linguistic indicators of LLM assistance, including preliminary 2024 observations. It does not measure hallucination prevalence or establish that all such assistance damages science.
- Publisher
- Andrew Gray / arXiv
- Kind of source
- Preprint, not peer reviewed
- Describes events of
- Public from
Where it was said (1)
Said once in the sources this register covers.
: Marcus on AI, The race between positive and negative applications of Generative AI is on – and not looking pretty
On the other hand, the use cases on the negative side of the ledger are proliferating fast, literally faster than I can report them here, from fake books (undermining authors) to rings of fake websites designed to sell ads
- Where
- paragraph 5
- Published
- 2024-03-29
- Source observed
- , live page
- Access
- full text, public
History
Assessment versions
- Assessment marcus-the-race-between-positive-and-negative-0c70cf-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-the-race-between-positive-and-negative-0c70cf-a1.