Pith. sign in

REVIEW 4 major objections 5 minor 4 references

Hype and Adoption of Generative Artificial Intelligence Applications

T0 review · 4 major / 5 minor · reviewed 2026-08-16 · deepseek-v4-flash

Pith's one-line read This paper claims that generative AI adoption follows both the Gartner Hype Cycle and the Kübler-Ross Change Curve, based on tweet sentiment and emotions over the first 100 days after ChatGPT's release.

desk verdict A well-written case study undone by a post hoc emotion-to-stage mapping that makes the Kübler-Ross validation circular and a sentiment curve that never goes negative. read the letter →

arxiv 2504.18081 v1 pith:4US6MMN7 submitted 2025-04-25 cs.CY

classification cs.CY
keywords generativeAIGartnerHypeCycleKübler-RossChangeCurvesentimentanalysisemotionTwitterdatatechnologyadoptionChatGPT
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

The paper claims that public reactions to generative AI, as seen in tweets from the first 100 days after ChatGPT's release, follow two classic models at once: the Gartner Hype Cycle, a five-stage model of technology expectations, and the Kübler-Ross Change Curve, a seven-stage model of emotional response to change. Using daily sentiment scores and fine-grained emotion classifications, the author reports that enthusiasm peaked around day 30, fell into disappointment, then stabilized at moderate optimism, matching the hype cycle. On the emotional side, early surprise, fear, and denial gave way to frustration, sadness, and finally neutrality and acceptance, matching the change curve. If right, the paper offers a compact, real-world demonstration that both models apply to generative AI and provides a timeline for businesses and policymakers to manage expectations and support adoption.

What carries the argument

The central machinery is the combination of two scoring pipelines applied to timestamped tweets. Sentiment is measured with VADER's compound score to produce a daily sentiment curve that is compared stage-by-stage to the Gartner Hype Cycle. Emotion is measured with EmoRoBERTa, which assigns each tweet to 28 fine-grained emotion categories; the paper assigns those emotions to seven Kübler-Ross stages by the month in which each emotion's score is highest (Table 1), then plots daily emotion scores as curves. The load-bearing step is that mapping: a post hoc assignment of, for example, surprise and joy to 'shock,' disgust and fear to 'denial,' and annoyance and confusion to 'frustration,' which turns independent emotion scores into a Kübler-Ross trajectory.

What would settle it

A direct test would pre-register the mapping between emotions and Kübler-Ross stages before data analysis, then apply the same pipeline to tweets about a later generative-AI release and check whether the predicted stage order appears. The claim would be undermined if, for example, surprise peaked after anger, or if the sentiment trough did not align with the peak of fear and confusion; more pointedly, since several emotions assigned to 'shock' and 'denial' both peak in month 1 and 'sadness' and 'decision' both peak in month 3, a re-analysis that reassigns those emotions to different stages would show how much of the Kübler-Ross fit is an artifact of the chosen mapping.

Watch

Extended reading notes

Core claim

On the paper's own terms, the central discovery is that the movement of sentiment scores closely follows the pattern predicted by the Gartner Hype Cycle, and that emotional reactions evolved from initial shock and denial to eventual acceptance and integration, so both stated hypotheses are effectively validated. The sentiment curve is bell-shaped: it rises to a peak of inflated expectations around day 30, dips to a trough of disillusionment near 0.24, then climbs to a plateau near 0.27, never turning strongly negative. The emotion curves, mapped from 28 categories to seven Kübler-Ross stages, show early peaks for surprise, joy, disgust, fear, and relief, followed by anger and confusion, then sadness and disappointment, and finally gratitude, curiosity, neutrality, and approval. The author interprets this as evidence that generative AI adoption is a dual-stage process, cognitive and emotional, and argues that the plateau of productivity sits higher than the starting point, reflecting genuine integration.

Load-bearing premise

The load-bearing premise is the post hoc mapping of 28 detected emotions to the seven Kübler-Ross stages based on the month in which each emotion peaked, a choice made after inspecting the data rather than derived from the change-curve literature; if that mapping is not valid, the emotion analysis provides no independent evidence for the Kübler-Ross order.

Editorial extensions

If this is right

  • Businesses and policymakers gain a predictable timeline: enthusiasm peaks around day 30, criticism troughs near day 60, and moderate optimism stabilizes by day 100.
  • Change-management teams can treat early resistance as expected and design training and communication for the frustration and sadness phases rather than treating them as product failures.
  • Because sentiment never turns strongly negative, the paper implies that generative AI is being integrated cautiously but successfully, so continued investment and rollout are justifiable.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • The paper's post hoc emotion-to-stage mapping is descriptive, not predictive; a stronger test would pre-register the mapping and apply it to a later AI release, such as a new model generation, to see if the stage order repeats.
  • The large positive slope for 'neutral' (2.22) suggests the main late-stage signal is not growing enthusiasm but growing habituation—users stop feeling strongly about the tool as it becomes background infrastructure, which is consistent with the plateau of productivity but also means the curve may be measuring attention fatigue, not satisfaction.
  • The dissociation between positive sentiment (never negative) and negative emotions (fear, disgust rising) hints that people can evaluate a tool as useful while still being affectively uneasy; future work could separate evaluative approval from emotional acceptance as distinct adoption barriers.
  • The 100-day window ends as the plateau begins; extending the data to a full year would reveal whether the plateau holds or whether a second hype cycle appears with new model releases, a test the paper's design leaves open.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

4 major / 5 minor

Summary. This paper analyzes tweets about ChatGPT, Bing AI, and Microsoft Office Copilot collected over the first 100 days after the public release of ChatGPT. The authors compute daily sentiment scores with VADER and daily emotion scores with EmoRoBERTa, then interpret the sentiment curve as evidence for the Gartner Hype Cycle and the month-by-month peak order of 28 emotions as evidence for the Kübler-Ross Change Curve. The paper concludes that both hypotheses are 'effectively validated.'

Significance. If the empirical link were established with rigorous methods, this study would be a valuable longitudinal complement to snapshot sentiment analyses and an interesting extension of the Gartner/Kübler-Ross frameworks to generative AI. The shift from information-seeking to content-creating users is a worthwhile framing. However, the current evidence is entirely descriptive: no inferential statistics, no pre-registered thresholds, and—most importantly—the Kübler-Ross stage mapping is constructed after inspecting the data, so the central validation claim cannot be accepted on the evidence presented.

major comments (4)
  1. [Section 4.2, Table 1] The assignment of the 28 emotions to Kübler-Ross stages is post hoc: each emotion is placed in the stage corresponding to its observed month of highest score, and this same ordering is then presented as evidence for the Kübler-Ross sequence. Because the stage labels are derived from the data, the observed progression from shock/denial to integration is true by construction and cannot independently validate Hypothesis 2; the paper needs an a priori mapping from EmoRoBERTa labels to stages or a separate confirmatory test. The overlap in the authors' own table (shock and denial both peak in month 1; sadness and decision both peak in month 3) further weakens the claimed monotone progression.
  2. [Section 4.1, Figure 3] The Gartner Hype Cycle interpretation is not supported by quantitative evidence: all reported sentiment scores are positive, the 'trough' is a drop from about 0.37 to 0.24 with no confidence intervals, no error bars, and no statistical test against a null or an alternative trajectory. The paper should report uncertainty, specify pre-registered thresholds for stage boundaries, and test whether the observed curve fits a hype-cycle shape better than a flat or monotone trend.
  3. [Section 3] The methodology omits essential reproducibility details: no dataset size, collection dates, exact search queries, language filters, deduplication procedure, tweet-level or aggregation-level unit of analysis, or details of how EmoRoBERTa scores were aggregated. Without these, and without code or data, independent verification is impossible. This also affects the validity of the slope values reported in Table 1.
  4. [Section 5] The statement that both hypotheses are 'effectively validated' is not supported by the preceding analyses, which are descriptive and contain no inferential statistics, significance tests, or effect sizes. The paper should temper its conclusions to 'consistent with' at most, or provide proper hypothesis tests.
minor comments (5)
  1. [Section 4.2, Table 1] The emotion numbering is inconsistent and the rows are not ordered by number, which makes the table hard to read.
  2. [Figure 3] The x-axis runs from -15 to 125 and the axes have no labels; the figure should be made self-contained.
  3. [Section 3] The sentiment tool is called 'VandeSentiment' but the reference is to VADER (Hutto & Gilbert, 2014); correct the name and capitalization.
  4. [Section 2.2] The abstract and method list experimentation and decision as Kübler-Ross stages, but the standard Kübler-Ross model has five stages; clarify that a seven-stage adaptation is being used.
  5. [Figure 2] Figure 2 is referenced but not described or explained in the text; add a caption and discussion.

Circularity Check

2 steps flagged · score 8.0 of 10

Kübler-Ross validation is circular: Table 1 assigns stages after seeing each emotion's peak month, then presents the resulting peak ordering as confirmation; the Hype Cycle fit is a post hoc relabeling of the sentiment curve.

  1. self definitional [Section 4.2, Table 1; Section 5]
    "Table 1 : Emotion trends Emotion Month of highest scores Slope of trendline Kübler-Ross stage 1 surprise 1 -1.05 Shock 2 joy 1 0.38 Shock 3 admiration 1 0.00 Shock 4 pride 1 0.34 Shock 5 disgust 1 0.47 Denial 7 embarrassment 1 0.32 Denial 8 fear 1 0.41 Denial 9 relief 1 0.79 Denial"

    The Kübler-Ross stage for each emotion is assigned from the observed 'Month of highest scores' column: emotions peaking in month 1 are labeled Shock or Denial, month 2 frustration, month 3 sadness/decision, etc. The same constructed ordering is then presented as evidence that 'users’ emotional reactions evolved from initial shock and denial to eventual acceptance and integration' and that Hypothesis 2 is 'effectively validated.' Because the stage labels are chosen post hoc from the peak months, the claimed Kübler-Ross sequence is true by construction and cannot independently confirm the model. The mapping also overlaps (shock and denial both peak in month 1; sadness and decision both peak in month 3), so even the constructed sequence is not the clean progression claimed.

  2. renaming known result [Section 4.1, Figure 3; Section 5]
    "The movement of the sentiment scores closely follows the pattern predicted by the Gartner Hype Cycle. The bell -shaped curve and subsequent leveling off reflect a typical trajectory of new technologies—from rapid excitement to disappointment, followed by gradual recovery and stabilization."

    The sentiment moving average is first described as rising to about 0.37 near day 30, dipping to about 0.24, and stabilizing near 0.27; the same plotted curve is then declared to 'closely follow' the Gartner Hype Cycle. There are no pre-registered thresholds for what counts as a peak, trough, or plateau, no null model, and no independent or out-of-sample test. The 'trough of disillusionment' is inferred from a modest dip that remains positive. Thus the validation of Hypothesis 1 is a post hoc relabeling of the observed trajectory as Gartner stages rather than a test of the model against data.

full rationale

The central circularity is in the Kübler-Ross validation. Table 1 is constructed by sorting each of the 28 EmoRoBERTa emotions according to the month in which its score peaks, then attaching Kübler-Ross stage labels (month 1 -> Shock/Denial; month 2 -> Frustration; month 3 -> Sadness/Decision; month 4 -> Experiment; months 4-5 -> Integration). Section 4.2 then reports that users 'evolved from initial shock and denial to eventual acceptance and integration' and Section 5 declares Hypothesis 2 'effectively validated.' The claimed emotional sequence is therefore true by construction: the observed peak months are the very criterion used to assign the stages, so the stage order cannot independently confirm the Kübler-Ross order. The mapping is also unstable (shock and denial both peak in month 1, sadness and decision both peak in month 3), so even the constructed sequence is not the clean monotone progression described. The Hype Cycle test is weaker but analogous: the sentiment curve is described as a bell-shaped peak, trough, and plateau after the fact, with no pre-specified thresholds or null model, so Hypothesis 1's 'validation' is a post hoc relabeling of the same moving average. The paper's self-citations (e.g., Truong & Hoang 2022; Truong 2023) are not load-bearing for this finding; the circularity comes from the table construction and post hoc curve reading, not from citation. Because the paper's distinctive theoretical claim—that generative AI adoption follows the Kübler-Ross Change Curve—reduces to a data-constructed mapping plus the same data presented as confirmation, the circularity score is high.

Assumptions & free parameters 4 free parameters · 5 assumptions · 0 invented entities

The paper rests on several domain assumptions: that the Hype Cycle and Change Curve generalize to generative AI, that tweets represent public opinion, and that the text-analysis tools are valid for this content. It also introduces an ad hoc rule that the month of an emotion's peak marks a Kübler-Ross stage boundary, and it never specifies the moving average window or exact search query, leaving multiple free choices.

free parameters (4)
  • Emotion-to-Kübler-Ross stage assignment = Table 1 mapping (28 emotions to 7 stages)
    The grouping is chosen after seeing each emotion's month of highest score, so it is a hand-fitted mapping rather than a theoretical one.
  • Sentiment peak and trough values = Peak 0.37, trough 0.24, plateau 0.27
    These stage markers are read by eye from the moving average curve; no formal criterion or uncertainty is reported.
  • Observation window = 100 days after ChatGPT release
    The window length is arbitrary; different lengths could change the shape of the curve and the inferred stages.
  • Moving average window = Not stated
    The paper refers to a moving average but never specifies its period, which affects the location of the claimed peak and trough.
assumptions (5)
  • domain assumption The Gartner Hype Cycle is a universal description of technology adoption.
    The paper assumes the five-stage model applies to generative AI without testing alternatives (Section 2.1).
  • domain assumption The Kübler-Ross Change Curve can be transferred from grief to technology adoption.
    The extension of a clinical grief model to user reactions to software is assumed rather than argued (Section 2.2).
  • domain assumption Tweets are a representative sample of public opinion on generative AI.
    The method treats unsolicited tweets as a faithful measure of public sentiment, without addressing platform bias, bots, or demographics (Section 3).
  • domain assumption VADER and EmoRoBERTa produce valid sentiment and emotion labels for this dataset.
    Neither model is validated on a sample of the collected tweets; no accuracy, agreement, or calibration metrics are reported (Section 3).
  • ad hoc to paper The month with the highest score for an emotion marks that emotion's Kübler-Ross stage.
    This operationalization is introduced for this analysis and is not justified; it produces overlapping stages, such as shock and denial both peaking in month 1 (Table 1).

how reviews work

0 comments
Cite this review

Pith. "Pith review of Hype and Adoption of Generative Artificial Intelligence Applications." pith.science (2026). https://pith.science/paper/4US6MMN7

@misc{pith2026250418081,
  author       = {Pith},
  title        = {Pith review of: Hype and Adoption of Generative Artificial Intelligence Applications},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/4US6MMN7}},
  note         = {Machine review of arXiv:2504.18081}
}
read the original abstract

New technologies create opportunities while displacing others. They enhance life by supporting entertainment, education, and social connectivity but also replace humans in productivity and analytical tasks. Adapting to these shifts requires technical adjustments and social readiness. For digital transformation to succeed, organizations and their workforce must be psychologically prepared. We are entering the era of Generative AI with tools like ChatGPT, Bing AI, and Microsoft Office Copilot. Understanding public sentiment toward these innovations is crucial for refining technology acceptance models and informing market strategies. Using the Gartner Hype Cycle and Kubler-Ross Change Curve, this study suggests that generative AI adoption is a dual-stage process. It follows the phases of technology trigger, peak of expectations, trough of disillusionment, slope of enlightenment, and plateau of productivity, while also reflecting emotional stages like shock, denial, and integration. The study used sentiment and emotion analysis on a large dataset of tweets about generative AI, translating them into scores to track user responses over time. Unlike prior research, which offered a snapshot of sentiment, this study captures the dynamic evolution of attitudes, linking empirical evidence with theoretical frameworks. It shifts the focus from information seekers to content creators. With the release of generative AI tools, there is a significant gap in understanding societal reception and adaptation. Policymakers face uncertainty about guiding markets for these changes. This research validates the applicability of the Gartner Hype Cycle and Kubler-Ross Change Curve to generative AI. It provides insights for businesses in integrating these tools and crafting policies to enhance readiness and resilience.

Figures

Figures reproduced from arXiv: 2504.18081 by the authors.

Figure 1
Figure 1. Gartner Hype Cycle (Fenn & Raskino, 2008) The Gartner Hype Cycle has proven to be a useful tool for explaining and predicting public reception of various technologies across sectors such as education, healthcare, governance, and industry. For example, innovations like virtual reality, IoT platforms, and 4D printing have all been mapped onto this model to understand where they stand in terms of public interest and ad… view at source ↗

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

4 extracted references · 3 canonical work pages

  1. [1]

    G., Bakombo, S., Konkle, A.T.M

    Abebe, R. G., Bakombo, S., Konkle, A.T.M. (2023). Understanding the Response of Canadians to the COVID -19 Pandemic Using the Kübler -Ross Model: Twitter Data Analysis. In (Vol. 20). Alhadlaq, A., & Alnuaim, A. (2023). A Twitter -Based Comparative Analysis of Emotions and Sentiments of Arab and Hispanic Football Fans. Applied Sciences, 13(11),

  2. [13]

    So what if ChatGPT wrote it?

    Bonebright, D. A. (2010). 40 years of storming: A historical review of tuckman’s model of small group development. In (Vol. 13). 17 Bridges. (2022). Bridges Transition Model - William Bridges Associates. In. Carr, N. (2017). A Mathematical Justification of the Gartner Hype Curve A Mathematical Formulation of the Emerging Risk Curve and Justification for t...

  3. [3036]

    https://doi.org/10.3390/app14073036 19 Williams, M., Braddock, Martin. (2019). AI Case Studies: Potential for Human Health, Space Exploration and Colonisation and a Proposed Superimposition of the Kubler -Ross Change Curve on the Hype Cycle. In (Vol. 8). Workshop, S. (2018). Satir Change Model. In. Zhou, L., Deng, X., Ning, Z., Zhao, H., Wei, J., & Leung,...

  4. [6729]

    AlQahtani, A. S. (2021). Product sentiment analysis for Amazon reviews. International Journal of Computer Science & Information Technology (IJCSIT) Vol,

Pith tools

Reviewed August 16, 2026 · model on record in the stance chip above.