Unkwn42 is the missed-literature check you run before you submit a climate or earth science paper. A plagiarism checker makes sure you didn't copy words; unkwn42 makes sure you didn't miss earlier work. You paste your abstract and reference list and it answers three questions: is my main claim already published, what earlier work am I missing, and is my reference list clean. Every paper cited is a real retrieved record, never generated.

How does Unkwn42 help researchers?

Before submitting, a researcher finds the papers a referee is likely to raise: earlier work making the same claim, work from a neighbouring field or under older terminology, and problems in the reference list such as retracted papers or preprints that have since been published. Before each check, the papers that cite the same sources are added, because that is where the missed paper usually is.

PhD students and postdocs in climate and earth science who are about to submit a journal paper. Later, the same check could run on every submission a journal receives.

Does Unkwn42 reject ideas that overlap with prior work?

No. It is not a pass or fail uniqueness test. Replication, extension, and doing a known study in a new place are legitimate research. The verdict has three states: Claimed (a specific paper already asserts the claim), Adjacent (the nearest work is distinguishable, with the delta stated), and Open (no paper unkwn42 holds asserts it). Even a Claimed result is information for a decision: what to match if replicating, or what to differentiate against if extending.

Will Unkwn42 invent or hallucinate citations?

No. Every paper cited is a real record retrieved from the literature, not generated. If the system cannot point to a specific paper, it does not make the claim. Grounding is verified in code before anything reaches you, and every result links to the source so you can check it yourself.

What do I need to give it?

Your abstract and your reference list, as pasted text, a .bib or .ris file, or a list of DOIs, and an email address for the sign-in link. The check takes about 10 minutes and emails you when it is ready.

Why not use a general AI research assistant or Google Scholar?

General research agents answer across all of science; Unkwn42 does one question in one field and goes deep. Keyword search also fails silently when you do not know the words: a paper that already did your work can sit fully indexed under a term your subfield stopped using, and no search you would think to run will surface it. Unkwn42 expands your idea across renamed and drifted terminology to retrieve exactly that prior work.

Your first check is free, one per email address: the answer on your main claim in full, how many earlier papers you are missing with the closest one in full, and how many reference problems were found. The full check is $19 per paper, with re-checks for 30 days. See https://www.unkwn42.com/pricing.

67 weeks to review the literature, and still unsure

Name: Unkwn42
Author: Unkwn42

A full systematic review takes over a year and keeps about 3% of what it finds. Most of us check prior work with far less method and a lot more doubt.

Rohit Pawar · 22 October 2026 · 3 min read · Updated 29 October 2026

Key findings

Across 195 completed systematic reviews registered in PROSPERO, the mean time from registration to publication was 67.3 weeks (Borah et al., 2017).
Those reviews' searches turned up anywhere from 27 to 92,020 studies, and on average 2.94% of what was found ended up in the review.
Checking a single idea for prior work often happens with no protocol at all, which is why finishing the search so rarely feels like knowing the answer.

During my dissertation I spent months finding papers. I don't think the months were the worst part. The worst part was that at the end of them I still wasn't sure I'd found the ones that mattered.

I wanted to know how that compares with doing it properly, so I looked for numbers on the most careful kind of literature search there is.

What a careful search costs

In 2017, Borah and colleagues looked at 195 systematic reviews of medical interventions that had been registered in PROSPERO, the international register for such reviews, and then published. On average it took 67.3 weeks to get from registration to publication. The reviews had a mean of five authors each.

Their searches turned up anywhere from 27 to 92,020 studies. On average, 2.94% of what a search found made it into the final review. Everything else, around 97 records in every 100, had to be looked at closely enough to be ruled out.

A systematic review is slow on purpose. There's a protocol written before the search starts, records are usually screened by more than one person, and every decision is written down so someone else can check it. It's the best answer we have to the question "what is already known about this?", and it takes well over a year from start to finish, with several people involved.

What most of us do instead

Checking whether an idea has been done, for a thesis chapter or a grant, often happens over an evening or two. A few searches on Google Scholar, a few keyword combinations, a skim of whatever comes back, a pile of open tabs. It's quicker than a systematic review by a long way, and it has none of the safeguards. Nobody writes down which words they tried, and nobody checks the field next door. The search ends because you got tired or the deadline arrived, not because you reached the bottom.

That's why finishing it so rarely feels like knowing. You haven't shown that the work doesn't exist. You've shown that the words you tried, in the database you used, didn't bring it up.

Where the doubt comes from

In my experience, most of it comes from vocabulary. The same idea gets different names in different fields and decades, so the search you ran in your own field's words can't reach the papers written in someone else's. I've written about two examples: marine heatwaves, which were often called warm anomalies before 2016, and yellow substance, which became coloured dissolved organic matter.

The rest comes from volume. If a careful review keeps about 3 in every 100 records it finds, the few that matter to you aren't necessarily on the first page, and a quick search is exactly the kind that stops there.

Asking the smaller question first

Most of the time you don't need a full review to decide whether to start. You need to know whether your specific claim has already been made, and what the nearest work is. That's a much smaller question, and it's worth answering before the big one, because the answer changes how you spend the rest of your reading.

That's the job I'm building unkwn42 for. You describe the idea in plain words, and it comes back with the closest existing work, including papers that use other words for the same thing, a line on how yours differs, and a note on how well we cover your area, so you know when to keep looking elsewhere. It doesn't replace a systematic review, and it isn't meant to. What it answers is the first question, the one that decides whether the weeks after it are well spent.

If you have a paper close to submission, run the check on it. The first one is free.

Sources

Borah R, Brown AW, Capers PL, Kaiser KA (2017). Analysis of the time and workers needed to conduct systematic reviews of medical interventions using data from the PROSPERO registry. BMJ Open 7(2): e012545. doi.org/10.1136/bmjopen-2016-012545