CiteOwl
CiteOwl
Blog/Getting citations right

Getting citations right

How to check a citation is real, why AI invents references, what to do when a source turns out fake, and how to cite in APA, MLA, and Chicago.

A citation does two jobs. It points a reader to the exact work behind your claim, and it shows you actually read that work. Both jobs break in the same place: the reference looks right, and the source either does not say what you claimed or does not exist at all.

That failure got much more common once students started drafting with chatbots. A language model predicts what a plausible reference looks like, so it produces real-sounding authors, a real journal, a sensible year, and a DOI with valid syntax that resolves to nothing. Published studies put the fabrication rate anywhere from a few percent to more than half, depending on the model and how it was asked.

The chain from a claim to a paper you can open

Every problem the twelve guides below deal with is a break in one chain, and the useful question is never "is this citation good" but "which link snapped". There are five, and each one fails in its own way.

  1. The claim. A sentence in your draft asserts something. Either it needs outside backing or it does not. Your own reasoning, your own arithmetic and genuinely common knowledge need none, and a citation bolted onto them is noise.
  2. The reference string. Authors, year, title, journal, DOI. This is the link a chatbot can produce out of nothing, because a reference is a pattern and patterns are exactly what a language model learned.
  3. The metadata record. The string either matches a record some publisher deposited or it matches nothing. This is the first link that can be checked mechanically, in one query, for free.
  4. The resolver. A DOI is not a web address, it is a key. doi.org looks it up and forwards you to wherever the publisher currently hosts the work, which is why a DOI survives a journal moving its website and a raw URL does not. A DOI that resolves proves a record exists. It proves nothing about what is written in the paper.
  5. The paper. You open it and read enough to confirm it supports your sentence. This is the link almost everyone skips and the one a marker actually tests, because it is the only one where reading is unavoidable.

Two things can still invalidate a citation after all five links hold: the paper was retracted after it was published, or the journal it appeared in is not doing the peer review it advertises. Both have their own register, and both are checkable in under a minute.

What each register holds, and what it cannot see

The registers behind links three, four and the two after-the-fact checks are public, and most of them answer an anonymous query with no account and no key. We ran those queries on 11 August 2026 and the counts below are what each register's own API returned that day, so treat them as a snapshot rather than a constant. Every one of them grows daily.

RegisterWhat it holdsCount on 11 August 2026What it cannot tell you
Crossref Metadata deposited by publishers: journal articles, book chapters, conference papers, preprints 185,346,103 records Whether the paper supports your sentence. A record is a filing, not a reading
DataCite DOIs for datasets, software, theses and repository deposits 133,322,923 DOIs The same. It also holds a lot of grey literature that was never peer reviewed
OpenAlex An open index built on Crossref plus repositories, with authors, venues and citation links 323,998,640 works Full text it has no open copy of. It knows the work exists, not what page 7 says
Retraction Watch Retractions, expressions of concern, corrections and reinstatements, hosted by Crossref 71,686 records, of which 66,197 are retractions Whether the paper is wrong. Retraction is an act by a publisher, and its reasons range from fraud to a duplicate submission
DOAJ Open access journals that passed its own vetting 23,303 journals Whether any particular article in one of them is any good

Do not read those four counts as a ranking. They describe overlapping and differently shaped universes: OpenAlex is larger than Crossref partly because it merges Crossref with repository and preprint records, so the same paper can arrive twice from different directions and be reconciled, and DataCite's DOIs are mostly not articles at all. What the table is for is the last column. Every register answers exactly one question, and a reference that clears all of them can still be a reference to a paper that does not say what you wrote.

One of these is newer than students realise. Retraction Watch spent years as a privately funded database; Crossref bought it in September 2023 and released the whole dataset publicly, which is the only reason checking a retraction is now a free lookup instead of a subscription. The counts above came from Crossref's REST API, DataCite's, OpenAlex's, the Retraction Watch dataset and DOAJ.

Where each guide plugs into the chain

Read in this order if you are working through a reference list you do not trust. Read the one that matches your link if you already know where it broke.

17 guides

All topics
Why AI makes up citations (and how often it happens)

Chatbots invent references that look real, down to the DOI. Why it happens, the fabrication rates measured in every published study from 2023 to 2026, and how to keep fake sources out of your work.

18 min read
How to check if a citation is real (in about five minutes)

Six checks with free tools that verify any reference in about five minutes, how to tell a fabricated citation from a garbled one, and what to do when one turns out fake.

13 min read
How to get ChatGPT to cite real sources

What actually cuts fabricated references in ChatGPT: turning on search, supplying your own sources, better prompts, and the one limit no prompt can fix.

12 min read
How to cite AI tools (ChatGPT, Claude, Gemini) in APA, MLA, and Chicago

The current APA, MLA, and Chicago formats for citing an AI tool, with a correct in-text and reference example for each, plus the bigger point: cite the real study, not the chatbot.

13 min read
How to avoid plagiarism (cite as you write, not after)

Most plagiarism isn't theft, it's drift: a paraphrase that stayed too close, a note you forgot was a quote. The habit that prevents it, and how to handle AI text honestly.

10 min read
ChatGPT cited a real paper that says something else

The source exists, the DOI resolves, and it still doesn't support your sentence. Why this failure is more common than fabrication, and how to catch it.

13 min read
ChatGPT gave me a DOI that doesn't work

A DOI can be perfectly formatted and still resolve to nothing. Five different causes, and the decision tree that tells them apart in three minutes.

8 min read
My professor says my sources don't exist. What now?

Audit your own reference list in the next hour, sort what you find, and reply with facts instead of excuses.

11 min read
How to tell if a journal article is fake

The article exists, but is the journal real scholarship? Predatory journals, hijacked clones, and paper mills, plus a five-minute check on any venue.

14 min read
Is the paper I cited retracted? How to check

How to check retraction status in seconds, how to read a retraction notice, and what to do when a source you already cited has been withdrawn.

12 min read
How common are fake AI citations?

One paper in 277 now carries a fabricated reference, up from one in 2,828 in 2023. What the audits measured, and why every number is a floor.

11 min read
The DOI links to the wrong paper

It resolves, the publisher is right, and the article is not yours. We measured how often one wrong digit still lands on a real paper.

12 min read

More on the blog

Every citation points at a paper you can open

Free to start. No card needed.

Start writing