The AI Integrity Crisis: Did the ‘Rat Penis’ Scandal Actually Dent Journal Submissions?

In February 2024, the scientific community was sent into a collective tailspin by an incident that, while bordering on the absurd, raised profound questions about the future of peer review. Frontiers in Cell and Developmental Biology, a prominent open-access journal, published a review article containing glaringly erroneous AI-generated figures. Among these was an anatomical diagram of a rat featuring a grotesquely oversized, non-biological organ, alongside a nonsensical signaling pathway map riddled with gibberish, pseudoscientific labels.

The incident became an overnight sensation, dubbed the “AI rat penis” scandal. It served as a visceral warning of the dangers inherent in the rapid, unmonitored integration of generative AI tools into academic publishing. However, two and a half years later, a deeper analytical look at the journal’s submission data suggests a counterintuitive reality: the scandal, despite its massive public fallout, may have had little to no impact on the journal’s long-term operational health.

The Anatomy of the Scandal: A Chronology

To understand the significance of this event, one must revisit the atmosphere of early 2024. Generative AI tools like Midjourney and DALL-E were becoming ubiquitous, and researchers were increasingly experimenting with them to create schematic illustrations for review papers.

• February 14, 2024: The paper, titled "Cellular functions of spermatogonial stem cells in relation to JAK/STAT signaling pathway," was published. Within hours, the image of the “AI rat” began circulating on X (formerly Twitter) and Reddit, sparking widespread ridicule.
• February 16, 2024: Faced with intense public scrutiny and a barrage of criticism regarding its editorial standards, Frontiers officially retracted the paper.
• Post-Retraction: The incident ignited a firestorm of discussion regarding the "publish or perish" culture, the reliance on automated systems for visual aids, and the erosion of the peer-review process. Frontiers, a publisher already under a microscope for its high-volume model, became the primary target for debates on scientific integrity.

Assessing the Damage: What the Data Says

For those who believed the scientific community would collectively "boycott" the journal, the data provides a sobering corrective. Using the R package PubMedLagR, analysts have mapped the publication and submission trends of the journal from 2015 through mid-2026.

The visual evidence is clear: while Frontiers in Cell and Developmental Biology experienced a noticeable decline in output, this trend began well before the February 2024 incident. The journal saw an explosive, pandemic-era surge in submissions—a phenomenon observed across many open-access journals during the COVID-19 lockdown—followed by a natural market correction.

When observing the monthly publication bar charts, the dotted line marking the “rat incident” shows no sharp, immediate drop-off. In fact, if anything, the journal’s publication numbers appear to have stabilized and even shown signs of recovery in the subsequent months. This suggests that the “scandal” was a blip on the radar for the average author, eclipsed by the broader macroeconomic and structural changes within the publishing industry.

Submission Dynamics

Critics might argue that publication numbers are a "lagging indicator"—that is, papers published in late 2024 were likely submitted well before the scandal broke. To address this, we must look at the "received dates" of papers.

Analysis of submission data confirms that the influx of new manuscripts did not suffer a catastrophic decline following the publicity of the AI blunder. The volume of incoming papers remained relatively consistent with the journal’s post-pandemic trajectory. Even if one accounts for potential fluctuations in rejection rates—which the current data does not explicitly isolate—there is no evidence of a mass exodus of authors. The journal’s workflow continued, suggesting that the “scandal” failed to register as a deterrent for researchers seeking a venue for their work.

The Drivers of Journal Selection

Why would a journal survive such a high-profile integrity failure? The answer lies in the harsh realities of modern academia. For most researchers, the selection of a journal is dictated by two primary factors:

1. The Journal Impact Factor (JIF): For many, this remains the gold standard for career advancement, grant applications, and tenure reviews. A high JIF often overrides concerns about a journal’s past reputational blunders.
2. Turnaround Time and Ease of Publication: In an environment where speed is essential for securing priority in discovery, the "hassle" of the submission process is a major deterrent. Journals that offer fast, streamlined review cycles gain a competitive edge, regardless of whether they have been the subject of controversy.

The irony here is palpable. The scientific community frequently argues that papers should be “judged on their own merits” rather than the prestige of the venue. Yet, by prioritizing impact factors and rapid publication, the community inadvertently creates a marketplace where high-volume, potentially high-risk journals remain the most attractive options, even after clear failures in oversight.

Official Responses and Editorial Reform

Following the 2024 incident, Frontiers—like many other major publishers—initiated a series of policy updates regarding the use of generative AI in manuscripts. These updates typically include:

• Mandatory Disclosure: Authors are now required to declare the use of AI tools in the creation of figures or the drafting of text.
• Human Oversight: Increased requirements for editorial staff to perform "sanity checks" on visual assets, moving away from a purely peer-based review for graphical content.
• Technological Audits: Investment in image-forensic software to detect AI-generated artifacts, similar to the tools currently used to detect image manipulation in western blots or microscopy.

While these measures are necessary, they are reactive. The core issue remains: the incentive structure of modern science heavily favors the publisher, and a temporary reputational hit is rarely enough to disrupt the established flow of submissions.

Broader Implications for Scientific Integrity

The persistence of Frontiers in the face of the “rat penis” incident serves as a case study for the fragility of reputation in the digital age. It suggests that while the internet is excellent at creating "virality" around scandals, this virality is ephemeral.

For the scientific community, the implications are twofold:

1. The "Scandal Fatigue" Phenomenon: Scientists, like the general public, are subjected to a constant stream of information. A scandal that seems monumental on social media today is often forgotten by the time a researcher is ready to submit their next paper six months later.
2. The Normalization of AI: As generative AI becomes integrated into every facet of research—from literature reviews to data analysis—the industry is moving from "fear of AI" to "management of AI." The 2024 incident may simply be viewed as a "growing pain" of a transition period, rather than a permanent stain on the journal’s record.

Ultimately, the data suggests that structural market pressures are significantly more influential than institutional reputation. As long as researchers are incentivized to chase impact factors and fast publication speeds, they will continue to submit to journals that provide these services, regardless of past missteps.

Conclusion

The “AI rat penis” incident remains a cautionary tale of what happens when the speed of technology outpaces the rigor of editorial oversight. However, the quantitative evidence demonstrates that science is a resilient, if not somewhat pragmatic, enterprise. The journal’s ability to weather the storm and maintain its submission numbers underscores the reality that for most, the convenience of the publication pipeline outweighs the weight of a single, albeit humorous, editorial failure.

As we move further into the era of AI-augmented research, the responsibility of maintaining integrity remains a shared burden. While publishers must tighten their controls, the community must also reckon with the incentives that prioritize volume over quality. If we continue to judge journals by their impact factors rather than their commitment to absolute transparency, we should not be surprised when history repeats itself.

Technical Methodology for Further Research

For those interested in verifying these trends or extending the analysis, the R code used to extract and visualize the publication data from PubMed is provided below. By utilizing the PubMedLagR package, researchers can perform longitudinal studies on journal performance and observe how external shocks—such as public controversies or policy changes—impact the scholarly communication ecosystem.

# Code snippet for analyzing journal publication trends
library(PubMedLagR)
library(ggplot2)

# Retrieve data for the journal in question
retrieve_journal_year_records("Front Cell Dev Biol", 2015:2026, batch_size = 200, papers_only = FALSE)
all <- pubmed_xmls_to_df(clean = FALSE)
all <- all[!duplicated(all),]

# Separate papers from other content types
retrieve_journal_year_records("Front Cell Dev Biol", 2015:2026, batch_size = 200)
pprs <- pubmed_xmls_to_df()
all$paper <- ifelse(all$pmid %in% unique(pprs$pmid), "paper", "other")
all$year_month <- as.Date(paste0(substr(all$pubdate,1,7),"-01"))

# Visualization of publication trends
ggplot(all, aes(x = year_month, fill = paper)) +
  geom_bar() +
  geom_vline(xintercept = as.POSIXct(as.Date("2024-02-14")), linetype=2) +
  labs(x = "Date", y = "Publications per month") +
  theme_classic()

Note: This analysis is based on available public data from PubMed. Trends in rejection rates or internal editorial filtering processes remain proprietary to the publisher and were not accounted for in this assessment.