“Generative AI does not do the things that it's being sold as doing, and the things it can actually do aren't the kind of things that create business returns, automate labor, or really do much more than one extension of a cloud software platform.”
Our reading, the sentence that was checked:
Generative AI does not do what it is marketed as doing, and what it can do neither creates business returns nor automates labor.
Assessment: Contradicted Evidence as of
Where and when Ed Zitron wrote it
Where's Your Ed At, The Hater's Guide To The AI Bubble, paragraph 303.
- Said
- Source observed
- live page, full text, public
- Evidence as of
- the cutoff for this assessment
- Assessment recorded
- assessment wyea-the-haters-gui-64c7a3-a1
We recorded 49 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 2025-07-21. 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
- Generative AI does not do the things that it's being sold as doing
- the things it can actually do aren't the kind of things that create business returns
- automate labor
- really do much more than one extension of a cloud software platform
Our interpretation
- Read as a present-tense claim about deployed generative AI capabilities, rather than a forecast.
- With no reporting period specified, use April–June 2025, the latest completed calendar quarter before publication.
- The plain reading denies meaningful business returns and labor automation. A documented deployment performing a substantial share of work previously requiring people can contradict the automation clause.
- Task automation does not require eliminating whole jobs. Handling inquiries does not establish that every inquiry was resolved satisfactorily.
- Business returns require benefits exceeding relevant costs. Usage, productivity, revenue growth and gross savings alone do not establish net returns.
- The marketing clause requires matching an identifiable advertised capability to evidence about its delivery. The cloud-platform comparison supplies no separate measurable threshold.
- The claim concerns general capabilities, not a named company's financial metric; metric_scope is null.
Counts as supported if: Representative deployment evidence shows that advertised capabilities were not delivered and that actual capabilities produced neither positive business returns nor substantial automation of human tasks.
Counts as contradicted if: Credible evidence within the target period documents substantial task automation or positive realized business returns from deployed generative AI. This contradicts the broad denial without establishing that all marketing promises were fulfilled.
Retrospective: these criteria were written after the outcome was public.
Current assessment
Was it true for the period described?
Outcome: Contradicted
The dated evidence goes against 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 broad denial of labor automation is contradicted on the stated task-level reading. In May 2025, CX Dive reported that Klarna's deployed assistant still handled two-thirds of customer inquiries, based on its spokesperson's account. That is substantial task automation, although it does not establish complete job replacement, satisfactory resolution of every inquiry, or net financial returns. The same report describes quality concerns and renewed human recruitment. METR also found slower work in one coding setting. These limits matter, but do not establish that generative AI cannot automate labor. The marketing and industry-wide returns clauses remain unproven either way.
Limitations
- The decisive deployment figure is company-reported through a news outlet, not independently measured.
- The result depends on reading the sentence as denying substantial task automation. A narrower claim about industry-wide profitability would remain unresolved.
- The assessment does not establish that generative AI generally delivers its marketing promises.
- Quality problems and the continuing need for human service limit what the automation figure demonstrates.
Evidence (3 items)
Goes against the claim: CX Dive, republished by Yahoo Finance,
The chatbot still handles two-thirds of all customer inquiries.
Why it matters here: The report attributes ongoing customer-service automation to Klarna spokesperson Clare Nordstrom, contradicting the categorical task-automation denial. It also reports renewed human recruitment and quality concerns; the volume figure is company-reported and does not measure net returns or successful resolution of every inquiry.
- Publisher
- CX Dive, republished by Yahoo Finance
- Kind of source
- News report
- Describes events of
- Public from
Context only: METR,
19% longer
Why it matters here: This randomized study supplies material evidence that useful productivity gains were not universal. June marks the end of its February–June study window; because that window extends beyond the target quarter, and the study does not test a specified marketing promise, it is context rather than decisive support.
- Publisher
- METR
- Kind of source
- Independent measurement
- Describes events of
- Public from
Context only: Klarna,
Klarna today announced its AI assistant powered by OpenAI.
Why it matters here: Identifies the assistant underlying the later deployment report as OpenAI-powered. Its launch-period performance and projected savings are outside the target period and are not used to establish the outcome.
- Publisher
- Klarna
- Kind of source
- Company disclosure: the company describing itself, not an independent finding
- Describes events of
- Public from
What would settle it
Independent target-period resolution logs and a complete accounting of deployment costs and benefits would strengthen the automation finding and test net returns. Assessing the marketing clause requires identifiable promises and matching evaluations.
Where it was said (1)
Said once in the sources this register covers.
: Where's Your Ed At, The Hater's Guide To The AI Bubble
Generative AI does not do the things that it's being sold as doing, and the things it can actually do aren't the kind of things that create business returns, automate labor, or really do much more than one extension of a cloud software platform.
- Where
- paragraph 303
- Published
- 2025-07-21
- Source observed
- , live page
- Access
- full text, public
History
Assessment versions
- Assessment wyea-the-haters-gui-64c7a3-a1: Current. Outcome: Contradicted. 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.