Issue Nº 003 · Research

Find sources in half the time

Perplexity surfaces citations in seconds instead of twenty minutes of keyword-guessing on Google Scholar. The catch: it will hand you a confident-sounding source that doesn't hold up. Here's how to use it without getting burned.

Every dissertation starts the same way: a blank search bar and a topic you can describe in a sentence but not in the three keywords Google Scholar actually wants. You type something, skim ten irrelevant abstracts, retype it, and twenty minutes later you have one half-decent paper and a headache. Perplexity skips that whole stage. Ask it the question in plain English and it comes back with an answer and the sources behind it, in the time it takes to read this sentence.

That speed is real, and it's why Perplexity has become the first stop for a lot of students doing early-stage research. But it changes what you're actually doing when you search. Google Scholar makes you find the paper. Perplexity makes you trust a summary of the paper — and a summary is only as good as the thing generating it.

Why it's fasterYou're asking a question, not guessing a keyword

Academic databases were built for people who already know the vocabulary of their field. If you don't know that your topic is called "algorithmic price discovery" rather than "how AI predicts stock prices," you'll search the wrong term and miss the right papers entirely. Perplexity doesn't need you to know the vocabulary. It reads your question, works out what you actually mean, and searches on your behalf — then shows its sources inline, not as a reading list you have to chase down separately.

Used well, that makes it a map of a topic you know nothing about yet, not a replacement for reading the territory.

Prompt of the week

"What are the main academic positions on [your topic]? Name the camps, give me a source for each, and tell me where the disagreement actually is."

This gets you the shape of a debate in one pass — who disagrees with whom, and why — instead of a flat list of papers that all sound like they agree.

The catchConfident isn't the same as correct

The Tow Center for Digital Journalism at Columbia tested eight AI search tools against known news articles in 2025 and found that misattributed or fabricated citations were common across almost all of them — tools confidently linked claims to sources that didn't say what the tool claimed, or didn't exist in the form cited. Perplexity performed better than most of the tools tested, but "better than the worst AI search tools" is a low bar for something going in your bibliography.

The failure mode is specific: the summary sounds right, the citation looks real, and the link sometimes even resolves to a real paper — just not one that supports the claim you're about to make. A marker who clicks through and finds that mismatch won't blame the AI. They'll blame you.

The habit that worksClick through, every time

Treat Perplexity as the fastest way to find out what to read, not as the last stop before you write it down. The twenty minutes it saves you at the start are real. Spend some of them back double-checking the three sources you're actually going to use.

✦ End of issue ✦

Get the next one in your inbox

One AI tool, one prompt, three quick wins, every week. Free, and under 3 minutes to read.

Subscribe Free →