For expert reviewers VOL. 02 / 2026

Reviewer Guidelines

Everything you need to evaluate AI-generated articles with accuracy, integrity, and impact. Your review is a public companion piece — not a private report.

A note before you begin

Thank you for joining Latent Scholar as a reviewer. Because AI-generated content cannot be revised by an author, your review serves a unique purpose: it is not a private report prompting revisions — it is a public companion piece that helps readers understand how much they can trust the article, and how to use it.

01What makes this different

Not traditional peer review

If you have reviewed for academic journals before, you’ll notice some key differences.

Traditional peer review
Latent Scholar review
Goal: decide accept / reject / revise
Goal: document strengths and weaknesses of AI-generated text
Private feedback to authors
Public review published alongside the article
Authors will revise based on feedback
No revision — you review an AI text with no potential for revision
Judge originality and contribution
Judge accuracy, reasoning, and reliability
Usually anonymous (double-blind)
Your choice: named or anonymous
Key mindset shift

You’re not gatekeeping — you’re creating a public record of what AI can and cannot do. Finding flaws is valuable data, not just criticism.

02Your role

A trusted expert annotator

Your task is to help readers understand four things:

1

How much they can trust this article — Is it accurate? Are there errors?

2

What the AI did well — What is solid, well-reasoned, or useful?

3

Where the AI failed — What is wrong, missing, hallucinated, or misleading?

4

How this content should be used — A good starting point? Useless? Risky?

You are asked to
You are not expected to
Audit the article’s accuracy, coherence, ethics, and usefulness
Request revisions or resubmission from an author
Explain its strengths and limitations in clear language
Provide detailed line editing or coaching
Advise readers on how (or whether) it can be used
Fix the paper — your role is to evaluate and contextualise
03Core principles

How to hold the work

Fairness & Objectivity

Review by the same scholarly standards used for human-authored work. Do not penalise or privilege the article because it was generated by AI.

Transparency & Accountability

Provide clear, evidence-based feedback. Your review forms part of a transparent record tracking the limits of AI-mediated scholarship over time.

Scientific Integrity

Scrutinise claims, methods, and interpretations. Identify where reasoning aligns with accepted knowledge and where it diverges or becomes speculative.

Tone and style

Professional & Respectful

Critique the text, not hypothetical authors.

Clear & Concrete

Explain what is wrong, why it matters, and how to interpret it.

Balanced

Note genuine strengths as well as limitations.

Accessible

Write for a broad academic audience, not only specialists.

04The review process

Step by step

1

Select an article

Browse available articles and choose one that matches your expertise.

Look at the discipline / field tag
Read the title and short description
Check which AI model generated it
Articles with fewer reviews are higher priority
2

Read the full article

Read carefully, as you would any paper in your field. As you read, take notes on:

Claims that seem accurate vs. questionable
Arguments that are well-constructed vs. weak
Statements that feel suspiciously confident
Citations that look real vs. potentially fabricated
Gaps in reasoning or missing counterarguments
3

Complete the review form

The form has two parts: structured ratings (seven dropdowns) and a written evaluation. You will provide:

Reviewer information: name, email, institution
Publication preference: named or anonymous
Overall evaluation: Meets / Needs Work / Below
Ratings for each of the seven criteria
Your comprehensive written assessment
Pro tip

Don’t try to verify every citation on first read. Get a sense of the overall quality first, then spot-check the most important references.

05Automated checks

What runs before you see it

Each submission passes through automated checks before it reaches you.

Cite-Ref Score

Verifies that in-text citations appear in the reference section and checks whether references are valid and accessible.

AI-Generated

Uses detection tools to identify whether the article appears to be purely AI-generated content.

Plagiarism

Scans for potential plagiarism across academic databases and web sources.

What this means for you

Automated checks handle basic verification, but can’t replace expert judgment. Your role is to catch what automation misses: subtle errors in reasoning, misrepresented sources, and disciplinary inaccuracies only a domain expert can identify.

06Evaluation criteria

Seven dimensions

You’ll rate the article on seven dimensions using dropdown menus.

01Accuracy & Validity
Does the article represent facts, concepts, data, and methods correctly?
Excellent / Strong
Core claims are factually accurate; no significant errors; represents the field correctly.
Satisfactory / Minor
Mostly accurate but with some errors, oversimplifications, or outdated information that don’t undermine the main argument.
Weak / Major
Significant factual errors, misrepresentations, or hallucinated content that seriously undermines credibility.
02Evidence & Citations
Are sources appropriate, verifiable, and properly used?
Excellent / Strong
Citations are real, relevant, and correctly attributed; references support the claims; format is consistent.
Satisfactory / Minor
Most citations valid but some are questionable, misattributed, or don’t fully support the claims.
Weak / Major
Multiple fabricated citations; sources don’t exist or say something different; evidence doesn’t support claims.
03Methodology / Approach
Is the approach appropriate? Are methods correctly described?
Excellent / Strong
Methodology is appropriate; methods correctly explained; limitations acknowledged.
Satisfactory / Minor
Reasonable but the description is incomplete, oversimplified, or missing key considerations.
Weak / Major
Inappropriate, incorrectly described, or fundamentally flawed; would not pass expert scrutiny.
04Reasoning & Argumentation
Is the logic coherent? Does the argument flow? Are there gaps?
Excellent / Strong
Arguments are logical and well-structured; counterarguments addressed; conclusions follow from evidence.
Satisfactory / Minor
Generally logical but with some gaps, unsupported leaps, or missing counterarguments.
Weak / Major
Logical fallacies; conclusions don’t follow; major gaps; internally contradictory.
05Structure & Clarity
Is the article well-organised, readable, and professionally written?
Excellent / Strong
Clear organisation; appropriate sections; readable prose; meets academic standards.
Satisfactory / Minor
Generally clear but some sections are awkwardly organised, too verbose, or unclear.
Weak / Major
Poorly organised; hard to follow; significant writing-quality issues.
06Originality & Insight
Does it offer useful synthesis, novel perspectives, or valuable insight?
Excellent / Strong
Valuable synthesis, fresh connections, or useful frameworks; informative to readers in the field.
Satisfactory / Minor
Competent summary but mostly derivative; limited new insight; reads like a textbook summary.
Weak / Major
No meaningful contribution; superficial treatment; reader would learn nothing useful.
07Ethics & Responsible Use
Are there ethical concerns, unsafe recommendations, or potential for misuse?
Excellent / Strong
No ethical concerns; claims appropriately hedged; no dangerous content.
Satisfactory / Minor
Some overconfident claims or missing caveats, but unlikely to cause harm.
Weak / Major
Dangerous advice, bias, or misinformation that could cause harm; seriously irresponsible.

Overall evaluation

Meets Standards
A competent reader could reasonably rely on this article as a starting point. Errors are minor or easily identified.
Needs Work
Has useful content but significant issues that require careful reading. Readers should verify key claims independently.
Below Standards
Fundamentally unreliable. Readers should not trust this without substantial independent verification.
07Spotting AI failures

Predictable failure modes

AI-generated academic content fails in recognizable ways. Here’s what to watch for.

1

Hallucinated citations

The most common and serious issue. AI will fabricate citations that look real but don’t exist. Spot-check 3–5 key citations with Google Scholar, Crossref, or the DOI resolver.

!Red flags
Author names that don’t appear in the field
Journal names that seem slightly wrong
Very recent dates (often “2023” or “2024”)
DOIs that don’t resolve
Titles unusually “perfect” for the claim being made
2

Confidently wrong statements

AI doesn’t express uncertainty well. It states falsehoods with the same confidence as truths.

!Red flags
Definitive statements about contested topics
Statistics or percentages without clear sources
Claims that “research shows” without citations
Historical claims that seem too neat or dramatic
3

Surface-level synthesis

AI often produces text that sounds sophisticated but lacks depth.

!Red flags
Paragraphs that could apply to many topics
Correct definitions but no real analysis
Listing perspectives without engaging the tensions
Conclusion that just restates the introduction
4

Logical gaps & non sequiturs

Arguments that seem to flow but don’t actually connect.

!Red flags
Conclusions that don’t follow from the evidence
Missing steps in the argument
Contradictions between different sections
Transition words that don’t connect logically
5

Outdated or incomplete information

AI models have knowledge cutoffs and may present old information as current.

!Red flags
References to “current” events that may not be
Missing recent developments in fast-moving fields
Superseded findings presented as consensus
Ignoring major replication failures or retractions
08Writing your review

The narrative is the most valuable part

Beyond the dropdown ratings, you’ll write a narrative review. Structure it in four moves.

Opening summary

2–3 sentences. What is this article about? What’s your overall assessment?

Strengths

1–3 points. What did the AI do well? What sections are accurate and useful?

Weaknesses

As many as relevant. What errors? What’s missing? What would mislead a non-expert?

Guidance for readers

1–2 sentences. How should someone use this article? What should they be cautious about?

Weak vs. strong reviews

×Weak review

“This article has some good points but also some errors. The citations seem mostly okay. Overall it’s a decent summary of the topic but could be better. I would rate it as needing work.”

Why it’s weak: Too vague. No specific examples. Doesn’t help readers know what to trust or avoid.
Strong review

“This review article attempts to synthesise research on cognitive load theory in online learning. The AI demonstrates solid understanding of core concepts and correctly summarises the foundational work.

However, I identified several significant issues:

1. The citation ‘Morrison & Chen (2022)’ appears to be fabricated — I could not locate it in any database.

2. The claim of ‘a 40% improvement in retention’ is not supported by the cited sources and seems invented.

3. The section on worked examples conflates element interactivity with task complexity in ways that would confuse readers.

The article could serve as a basic introduction, but readers should independently verify all citations and treat specific statistics with scepticism.”

Why it’s strong: Specific examples. Identifies strengths and weaknesses. Tells readers exactly what to watch for. Creates useful benchmark data.
09Time management

The 30-minute review

You don’t need to spend hours. Here’s an efficient approach.

5 min
Skim & orient
Read abstract, headings, and conclusion. Get the overall structure.
12 min
Careful read
Read the full article. Take quick notes on issues as you go.
5 min
Spot-check citations
Verify 3–5 key citations. Note any that fail.
8 min
Write review
Complete the form. Select ratings, write your assessment.
Remember

You don’t need to find everything. Your goal is a useful expert perspective, not a comprehensive audit of every claim. Document what you find.

10FAQ

Frequently asked questions

What if I’m not sure if something is wrong?

Say so. “I couldn’t verify this claim” or “this seems inconsistent with my understanding” are perfectly valid. You don’t need to be certain — your professional judgment is the point.

Should I be harsh or generous?

Be accurate. Don’t inflate ratings to be nice, and don’t deflate them to seem rigorous. Finding problems is valuable data, not a criticism of anyone.

What if the article is actually pretty good?

That’s also valuable data. Documenting what AI does well is just as important as documenting failures.

Should I comment on the writing style?

Yes, if notable. AI-generated text often has a distinctive voice — overly formal, repetitive, or suspiciously perfect flow. Worth noting.

Can I review outside my exact specialty?

Adjacent areas are fine. A social psychologist can review a cognitive psychology paper — but don’t review organic chemistry if you’re a historian.

Named or anonymous?

Your choice. Named reviews get attribution (good for CV). Anonymous reviews are fine if you prefer privacy. Either is equally valued.

What if I disagree with another reviewer?

That’s fine and expected. Different experts notice different things. Both reviews add value — you don’t need to reconcile them.

How do I get CV credit?

List it under “Professional Service” or “Peer Review.” If your review is published with attribution, you can cite it directly.

11Quick checklist

Quick reference

Before you start
Is this article in my area of expertise?
Do I have 20–30 minutes to complete this?
Have I noted the AI model and article type?
While reading
Am I noting claims that seem questionable?
Am I watching for overly confident statements?
Am I tracking citations that need verification?
Am I noting logical gaps or contradictions?
Citation check
Have I spot-checked 3–5 key citations?
Did I check if cited papers actually exist?
Did I verify claims match what sources say?
Writing the review
Have I given specific examples, not vague comments?
Have I noted both strengths and weaknesses?
Have I given readers guidance on how to use it?
Is my overall rating consistent with my comments?
Before submitting
Have I completed all seven rating dropdowns?
Have I selected my overall evaluation?
Have I chosen my publication preference?
Have I agreed to the Reviewer Agreement?
Why it matters

Your review contributes to a living record of how AI models perform in scholarly reasoning across disciplines:

Benchmark AI models against human expert expectations
Build a transparent reference base for detecting AI-assisted writing
Encourage responsible, verifiable integration of AI in academic work
Questions or issues?

If you run into problems or have questions, reach out any time and we’ll get back to you — usually within 24–48 hours.

Contact us
Ready to begin

Pick a discipline. Dive in. Share what you find.

Your expertise is building the foundation for understanding what AI-assisted scholarship can — and cannot — reliably produce.