Peer review crisis - Featured

The Peer Review Crisis: A Deeper Insight

by

Table of Contents

Introduction

Peer review has long been upheld as the cornerstone of academic integrity—the mechanism by which scholarship earns its badge of legitimacy. For centuries, this process has served as the gatekeeper of knowledge, distinguishing rigorously vetted research from the merely speculative. Yet, as the scholarly publishing ecosystem grows more commercialized and digitized, peer review finds itself in the midst of an escalating identity crisis.

This isn’t a theoretical problem. It’s a structural one. Reviewer fatigue is at an all-time high, submission volumes have exploded, and fraudulent practices are on the rise. Artificial intelligence, while offering support, is also blurring the line between assistance and automation. Institutions continue to benefit from a system built on unpaid, invisible labor.

The peer review crisis is not just a procedural issue; it’s an existential challenge. If left unaddressed, it risks eroding the very credibility of academic publishing. The write-up unpacks the origins and drivers of this crisis, surveys the current state of journal publishing, and explores pragmatic, forward-thinking solutions for a system in desperate need of reform.

The System Was Never Perfect, But It Worked

The traditional peer review process has always been more ritual than science. Modeled loosely after 18th-century philosophical societies, the earliest versions of peer review involved a small circle of elite thinkers evaluating each other’s work. The process was informal, rooted in trust, and inherently biased—but it was good enough for a slower, smaller academic world.

As research became more institutionalized and journals proliferated, peer review took on a more codified form. Double-blind reviewing emerged as a way to mask identities, reduce bias, and level the playing field. Review reports became essential editorial tools, used not just to judge merit but to improve clarity and relevance.

Despite its flaws—subjectivity, delays, opacity—the system generally worked. Scholars reviewed each other’s work in the spirit of mutual advancement. And there was still enough slack in the system to allow for thorough reviews, careful edits, and thoughtful revisions.

That era is gone. The growth in global research output and the explosion of open access publishing have transformed what was once a collegial process into an industrial workflow.

Submission Overload: The Deluge No One Ordered

In 2024, over 5 million scholarly articles were published, surging by multiple folds in a decade. This surge is fueled by multiple forces: increased research funding, the rise of global universities, performance-based assessments, and tenure systems that equate publication count with academic value.

Publishers, eager to expand market share and maximize APCs (Article Processing Charges), are launching new journals at an unprecedented pace. MDPI, for instance, now publishes over 400 open access journals, many with aggressive publication schedules. Elsevier, Wiley, and Springer Nature aren’t far behind.

This ever-growing submission volume overwhelms editorial systems. A single editor may handle hundreds of papers per year, often without the support of a managing team. Finding willing reviewers becomes a logistical nightmare. Journals report rejection rates of over 80% for reviewer invitations. Some editors now resort to mass-emailing dozens of potential reviewers, hoping a few will say yes.

This “reviewer drought” has cascading effects: delays in decision-making, rushed or superficial reviews, and, in some cases, acceptance without adequate scrutiny.

Reviewer Fatigue: An Epidemic of Indifference

Peer review depends on one crucial assumption: that scholars will voluntarily give up their time to critique others’ work. But this assumption is fraying fast.

A typical reviewer invests 4 to 6 hours per manuscript. Some reports can take even longer, particularly in fields that demand replication or data validation. Yet, the reward is negligible. No payment. No significant career advantage. No guarantee the feedback will even be taken seriously.

A report by Publons found that the top 10% of reviewers globally are responsible for over half of all reviews. These individuals—often mid- to senior-career scholars—are inundated with requests and operate under constant deadline pressure. Many are burned out. Others have grown cynical, churning out superficial reviews just to get it over with.

Junior researchers are underrepresented in review panels, either because editors don’t trust them or because they lack formal training. This imbalance adds yet another layer of stress to the system.

If peer review is a communal responsibility, then the community is clearly in crisis. And the exhaustion isn’t just academic—it’s emotional, institutional, and systemic.

The Rise of “Fake” Peer Review

What happens when desperation meets a broken system? Fraud.

Over the last decade, peer review fraud has evolved from an occasional scandal to a widespread pattern of misconduct. Authors have been caught submitting fake reviewer names and email addresses, allowing them to “review” their own work. Some have created elaborate review rings—networks of fake identities that bounce favorable reviews between associated authors.

According to Retraction Watch search, more than 1,000 articles were retracted between 2014 and 2023 for manipulated peer review. Many of these cases involved journals with lax editorial oversight or automated reviewer selection processes.

The use of generative AI tools is making the situation even murkier. Tools like ChatGPT are generating entire review reports—sometimes more articulate than their human-written counterparts—and accepting them without scrutiny.

These aren’t just a few bad actors gaming the system. It’s a systemic vulnerability in a publishing model that rewards volume, speed, and appearance over authenticity.

AI Enters the Fray: Savior or Saboteur?

AI’s growing presence in scholarly publishing is a double-edged sword. On one hand, machine learning can streamline editorial workflows: flagging plagiarism, checking statistical methods, suggesting reviewers, or summarizing long submissions. On the other, it enables a new kind of intellectual laziness—outsourcing not just writing but also reviewing to algorithms.

Some reviewers, overwhelmed by requests, are now using AI to generate draft reviews. These are then lightly edited—or sometimes not at all—before submission. Editors, under pressure to move papers along, often don’t have time to vet these reports deeply. The result? Reviews that look coherent but may lack substance or contextual understanding.

There is a larger philosophical issue at play here: Can an AI, lacking domain expertise or ethical accountability, serve as a peer in “peer review”? And what does it mean when the “review” part is mechanized, but the “peer” part is not?

Journals need to draw clear boundaries. AI should assist, not replace, human judgment. Used properly, it can help flag weak submissions, detect patterns of fraud, and even identify underused reviewers. But giving it editorial authority is a shortcut that leads nowhere good.

Transparency: What’s Taking So Long?

Opaque processes breed distrust. Yet, the majority of peer review systems remain hidden behind closed doors. Reviewers remain anonymous, reports are kept confidential, and editorial decisions often come with little or no explanation.

This lack of transparency has real consequences. It enables editorial favoritism, allows implicit biases to go unchecked, and leaves authors confused or demoralized. Worse, it erodes public trust in science, especially in times of crisis like the COVID-19 pandemic, when bad studies can go viral before they’re retracted.

Open peer review, where reviewer names and reports are published alongside articles, is gaining traction. Journals like F1000Research, eLife, and BMJ Open have adopted more transparent models, inviting post-publication commentary and dialogue.

But resistance remains. Reviewers fear retaliation. Editors worry about administrative overhead. And legacy journals, comfortable in their prestige, have little incentive to change.

Still, transparency is the future. It’s not about eliminating anonymity altogether—it’s about accountability. When reviewers know their work will be seen, they’re more likely to be thoughtful and constructive.

The Economics of the Crisis

Let’s talk money—or rather, the lack of it. Peer review is the backbone of academic publishing, a billion-dollar industry. Yet, it is built almost entirely on unpaid labor.

Publishers charge thousands in APCs (Article Processing Charges) or subscription fees. Authors pay to publish, readers pay to access, and reviewers get… a thank-you email, maybe. It’s a business model that would be unthinkable in any other industry.

A 2021 study estimated that unpaid peer review labor represents approximately $1.9 billion annually in unreimbursed contributions to scholarly publishing, with over 15 million hours contributed. Universities absorb these costs as staff time but rarely acknowledge them in budgets.

This model is not just exploitative—it’s unsustainable. It incentivizes shortcuts, breeds resentment, and diminishes the perceived value of peer review. Some journals have experimented with small honoraria or discounts on APCs for reviewers. Others offer reviewer recognition platforms like ORCID or ReviewerCredits.

But until academic institutions formally recognize reviewing as a professional activity—with credits, funding, or promotion points—the imbalance will persist.

Disciplinary Disparities and Equity Gaps

The peer review crisis is not evenly distributed. Fields like physics and computer science, which embraced preprints early on, are somewhat buffered from the review crunch. Their culture values rapid feedback and open commentary.

Biomedical and social sciences, by contrast, are drowning in submissions. The pressure to publish in high-impact journals is intense. The result is a bottleneck: overworked reviewers, excessive desk rejections, and editorial decisions driven more by trend-chasing than academic merit.

Meanwhile, scholars from the Global South face unique barriers. Despite representing a growing share of global research output, they remain underrepresented in editorial boards and review panels. Language, time zones, and institutional hierarchies conspire to keep them at the margins.

Equity in peer review must be about more than inclusion. It requires structural reform: multilingual review platforms, better reviewer training, and active outreach to underrepresented communities. Diverse voices enrich science. But only if they’re given a seat at the table.

Peer Review Alternatives: Brave New World?

With the traditional model buckling, alternatives are gaining ground. None are perfect, but all represent attempts to rethink the fundamentals:

1. Preprints and Post-publication Review
Platforms like arXiv, bioRxiv, and medRxiv allow rapid dissemination of research before formal peer review. Post-publication platforms like PubPeer offer real-time community feedback. While these systems speed up knowledge sharing, they also risk amplifying unvetted or low-quality work if the moderation is weak.

2. Overlay Journals
These are journals that “overlay” peer review onto preprint platforms, selecting and reviewing papers already available online. Examples include Quantum and Discrete Analysis. They reduce redundancy and increase transparency.

3. Portable Peer Review
Initiatives like Review Commons allow peer reviews to follow a manuscript from one journal to another. This saves time and reduces reviewer workload. However, it requires coordination and buy-in across publishers—something that’s still rare.

4. Collaborative Review
Some journals encourage reviewers to discuss their reports before submission. This leads to more consistent feedback and fewer contradictory suggestions. But it demands more time and editorial resources.

Experimentation is healthy. But the goal shouldn’t be to replace peer review—it should be to fix its broken incentives and reframe it as a service to science, not just a checkpoint in publishing.

What Journals Can Do Now

Fixing peer review doesn’t require a revolution, but it does require courage. Journals should:

  • Expand and diversify reviewer databases.
  • Use AI tools to assist in reviewer matching, not replace human judgment.
  • Require transparency in editorial decision-making.
  • Introduce formal reviewer training for early-career researchers.
  • Recognize reviewing in tenure and promotion evaluations.
  • Provide tangible rewards, including modest honoraria, APC discounts, or publication credits.

Critically, journals must be willing to say “no” to speed for the sake of integrity. Faster isn’t always better, especially when it compromises the quality of the scholarship.

Conclusion

The peer review crisis is real, but it is not insurmountable. It reflects deeper tensions in academic publishing—between profit and purpose, speed and scrutiny, tradition and innovation.

Peer review remains essential. Not because it’s perfect, but because no viable alternative offers the same mix of accountability, expertise, and rigor. But the system must evolve. It must embrace transparency, support its participants, and treat peer review as professional labor, not volunteer charity.

If journals want to preserve their role as arbiters of credible science, then the time to reform peer review is not tomorrow. It’s now.

Leave a comment