SIFT In Action

Let's see what SIFT looks like in practice.

Imagine you're working on a university assignment about healthy aging. You start by asking a generative AI tool: "How does coffee affect human aging?"

The AI gives you a source titled: "Coffee doesn't just wake you up - it may help protect your body from aging." The article is published by Texas A&M University.

At first glance, this looks pretty convincing. It's from a university, it discusses scientific research, and it points to a published study.

But before using this information in your assignment, let's apply SIFT.

STOP

First, stop. Before we start copying information or citing the article, we need to think about what we're looking at.

The claim is that coffee may help protect the body from aging. The resource is a university news article.

Now ask yourself: What am I actually looking for?

If this is a university assignment, maybe you're looking for a scholarly source to cite. Or maybe you're looking for background information that will help you understand your topic.

That distinction matters because a source can be useful without necessarily being the best source for your assignment.

So we won't automatically trust this article just because the publisher is a university. Let's investigate.

INVESTIGATE

Instead of staying on the article page, open another browser tab and start looking around.

First: Who created the information?

The article summarizes research from Texas A&M University's College of Veterinary Medicine and Biomedical Sciences, and the author is a research professional who specializes in communicating research.

Next: When was it published?

The article was published in April 2026. For a topic involving current health research, that date matters because scientific understanding can change over time.

Next: Where did the information come from?

The article points to research published in the journal Nutrients. That's useful because now we have a trail we can follow.

Next: What kind of resource is this?

It's not the original research study. It's a university news article explaining research findings to a general audience.

And finally: Why was it created?

Its purpose is educational, but it is also highlighting research to the public. That doesn't make the source unreliable. It simply gives us more context about what we're reading.

Now that we understand the source better, let's look beyond it.

FIND BETTER COVERAGE

The next step is to find better coverage. Rather than relying on this one article, we want to see what other credible sources have to say about the claim.

The article points us toward a study published in Nutrients. Let's open that study.

We can also use Omni, the University of Guelph Library's search tool, to find the article and look for additional research.

For example, we might search for:

- coffee aging
- coffee longevity
- NR4A1

Now we can compare multiple sources.

Are researchers reporting similar findings? Are there different interpretations? Is there agreement about what the evidence actually shows?

The goal isn't to find information that confirms what we already believe. The goal is to understand the broader evidence and identify where credible sources agree or disagree.

Finally, we want to trace the claim back to its original context.

Let's return to the original article. The article presents the research in a simplified way for a general audience. But if we're going to use this information in academic work, we may want to go back to the original study.

What did the researchers actually investigate? What methods did they use? What did they find? And what limitations did they identify?

This step is especially important with AI-generated information. An AI tool may point us toward a real source, but that doesn't necessarily mean the source says exactly what the AI claims.

By comparing the original claim with the original research, we can determine whether the evidence actually supports the statement we're considering using.

And that's SIFT in action.

We stopped before accepting the information. We investigated the source. We found better coverage. And we traced the claim back to its original context.

Notice that we didn't have to decide immediately whether the original article was 'good' or 'bad.' Instead, we gathered enough information to make an informed decision.

As a University of Guelph student, Omni can be one of your most useful tools in this process. You can use it to find scholarly articles, books, newspapers, and other library resources when you're looking for better coverage or tracing a claim back to its original source.

Generative AI can be useful for discovering information and identifying potential sources. But finding a source is only the beginning.

Before you use information in your academic work, remember to SIFT:

Stop.

Investigate.

Find better coverage.

Trace claims back to their original context.

In an age of generative AI, strong research isn't just about finding information. It's about verifying it.

[Questions? Contact us. lib.uoguelph.ca] 

[Music by Google Flow] 

[This work is licensed under  CC-BY-NC-SA 4.0 , except as otherwise noted.]