Bring Sanity to the scientific record.
When a journal retracts a paper, the paper should stop counting. It doesn’t. I traced every retracted paper with a DOI: 32,626 of 51,414 were cited again after they were pulled.
Part of the SaneApps family
Retracted papers keep getting cited.
Hardly anyone reads a retraction notice. In one large study of biomedicine, only 5.4% of sentences citing a paper after its retraction said it had been retracted.1
of retracted papers were cited again after the retraction: 32,626 of 51,414.
citations, in all, to papers that had already been retracted.
retracted papers were still being cited in 2025 or 2026.
Source: Retraction Watch database via Crossref, cross-checked with the OpenAlex citation graph. Updated 16 Oct 2026, 01:42 UTC. What these numbers don’t say →
Retracted2 Feb 2010
Ileal-lymphoid-nodular hyperplasia, non-specific colitis, and pervasive developmental disorder in children
The paper behind the MMR vaccine scare.
OpenAlex yearly counts start in 2012. *2026 is year to date.
Many of these cite the paper to warn about it. Some may still rely on it. A count can’t tell which, so Phase 2 reads every one.
Show the numbers
| Year | Citations |
|---|
Then I read 55,688 of those citations.
Every citation made after retraction to a PubMed-indexed paper, wherever the citing paper is open access. Three AI models read each citing passage; on a blind sample they matched a separate AI reader 94.7% of the time. A human check comes next, so these figures are provisional.
cited the retracted paper as support.
flagged it as retracted or unreliable.
still cited it as support when their own reference list said “retracted”.
The 40.7% and 3.3% are shares of the 51,626 citations the panel could judge: the 55,688 read, minus citations to administrative retractions (out of date, replaced, moved journal — reported apart) and 18 it could not call. Support fades slowly: 43.6% in the first two years after retraction, 34.9% after ten. Papers retracted for misconduct were cited as support as often as honest errors. Method, limits and data → · Search every citation →
Two things stand in the way.
A machine reads it all. You can check the machine.
1Cheap enough to read everything.
Reading is now almost free: the census read 55,688 citations for about $30. People still check the machine, and that is where the money goes.
| Reading 55,688 citations, three models eachmeasured, Phase 2 census, 13–14 Oct 2026 | about $30 |
| Every post-retraction citation in Phase 1262,998 citations at the same rate, if all were open | under $200 |
| Experts checking the machine’s flagsthe largest cost, by about ten to one | most of it |
| Campaign One, all three phases | under $50,000 |
Already run on real papers. The Phase 2 census read 55,688 citations of retracted papers and published every call, with its passage, for anyone to check.
2Open, so you don’t have to trust me.
A citation is a claim about a source. SaneCite reads the source, and every verdict quotes the passage it rests on.
The method is published before each run. The raw results come out with the findings, and all the code behind Campaign One is open source. If you can’t re-run it, I don’t publish it.
SaneCite was first proven in security compliance, where every answer has to trace back to approved evidence. The same engine now checks the scientific record. All the Campaign One code (the Phase 1 count, the Phase 2 reading panel and the model calls) and data are public under the MIT license, on GitHub.
The retraction audit, in three phases.
Every retracted paper is in the set because it was retracted, not because I picked it. Nobody gets singled out.
- Done
1Count
Match every retracted paper to every citation it got after the retraction. Mechanical, no judgement calls.
Out: the numbers above. Tables by journal and field, and the full dataset, publish next.
- Census read · human check next
2Read
For every citing paper with open full text, read the passage that cites the retracted work. Did the author use it as support, just mention it, or flag it? Experts check a random sample of the machine’s calls.
Census: 40.7% of 51,626 citations cited the retracted paper as support. Next: experts check a random sample.
- After Phase 2
3Tell
Send each journal that published a citation resting on retracted work the evidence. Then publish who corrected, who replied, and who said nothing.
Out: a public ledger of replies.
What a count does not tell you.
- A flag is not a finding.Some papers cite a retracted paper to criticise it. Phase 1 can’t see the difference. The Phase 2 census can: 40.7% cited it as support, and most of the rest mentioned it in passing.
- Authors often couldn’t know.The notice never reached them. I measure that failure, of the system, not of the scientists.
- Paperwork retractions are left out.Some retractions find no fault in the science: a review withdrawn as out of date, a temporary removal, a publisher’s error, a paper replaced at the same address. Since 16 Oct 2026 the count leaves out 1,858 such papers and their 23,251 citations, the same rule the Phase 2 census always used. The first count, 286,050, included them.
- I count by year, so I undercount.Only citations in the years after the retraction year are counted. Citations in the same year are left out.
Coverage: 53,272 of the 53,685 retracted works with DOIs, retracted in 2024 or earlier, matched in OpenAlex (16 Oct 2026); 51,414 after leaving out paperwork retractions. Retraction Watch holds 67,815 retraction records in all (15 Oct 2026). Not every counted citation is real: OpenAlex sometimes links the wrong paper, counts a retraction notice, or lists one paper twice. I checked 200 counted citations at random against the citing papers’ own reference lists (16 Oct 2026): about 96% were real, so the true total is probably about 254,000 (likely between 247,000 and 260,000).2
Every check, in the order it ran.
No highlights reel. These are six checks in a row, as they came out of the run. Most retracted papers are small and obscure, and many rows are clean.
| Loading the latest checks… |
Open the full feed → · Data: feed.json
Power
You can verify every number yourself. The data, the method and all the code behind Campaign One are public under the MIT license.
Love
Built to repair the record, not to shame people. Free for anyone who needs it, with the code and data open.
Sound Mind
Flags, not verdicts. No outrage, no villains. Just the count, the evidence and the limits.
- Whole sets only. I never pick targets.
- Method published before the run. Raw results published with it.
- A flag is not an accusation. A citation error is never called fraud.
- Anyone named gets to reply. The reply is attached to the record.
- No funder pays more than about a quarter of a disputed campaign.
- A campaign that misses its finish twice loses its funding.
- My own numbers get checked too. Every correction goes at the top of the page.
Pay for the reading.
Phase 1 counted. Phase 2 reads every citing sentence and has experts check the machine. Your money pays for people, and every spend goes on one public ledger.
- Hsiao T-K, Schneider J. Continued use of retracted papers: temporal trends in citations and (lack of) awareness of retractions shown in citation contexts in biomedicine. Quantitative Science Studies 2022. 722 of 13,252 post-retraction citation contexts in PubMed Central acknowledged the retraction.
- SaneCite Campaign One, Phase 1: first run 13 Oct 2026; corrected count 16 Oct 2026 (paperwork retractions left out). Retraction Watch database via Crossref (67,815 retraction records, 15 Oct 2026). Citations from OpenAlex counts_by_year; a citation counts as post-retraction if it falls in a calendar year after the retraction year. Exhibit: OpenAlex W2117847125, 2012–2026.