I spoke today about fact-checking at the 2025 national conference of ACES: The Society for Editing.

The theme was answering the audience’s questions about fact-checking, since often I present a lot of helpful tips and methods of fact-checking and verifying information, but I don’t leave time for many questions.

I got a lot of good ones … and four people stayed after the session for about a half hour to talk more about the subject. That certainly made me happy!

Yes, I’m a fact-checking nerd. I love it when people I know through ACES email me with a “I can’t find out anything about this” question and I can direct them to the right places. It’s especially fun to actually find the answer to a hard-to-figure-out query.

I got some good questions today. These slides have two of the questions posed on the app before the session began, and I worked up some resources. One is timely; it’s about data resources that might be disappearing right now from the web because they are part of government websites. I can’t say some information won’t just disappear, but there are groups working to make sure that doesn’t happen too often.

This next slide show is the original presentation, which is available to people who attended the conference on the conference app. But I’m all about sharing with anyone else who might need these tips.

If you have questions about fact-checking, find me on Threads, LinkedIn or Bluesky or email me through the contact form here.

By the way, the question of using AI for fact-checking did come up. Here’s my quick take on that: It’s a tool like a lot of other tools. It can’t do all the work for you, and it needs to be used correctly.

Consider these questions:

1. Can AI do fact-checking? Yes. Does it make mistakes? Yes. Does it miss things? Yes.
2. Are you sure you want to run an entire unpublished piece of writing through an LLM AI chatbot to have the chatbot check it? I’d be wary, because many times you are giving that system a right to that information, and you are losing your control over it.
3. If an AI chatbot can do a better job than you, then why do we need you?
4. Can an AI fact-check be trusted if it doesn’t provide sources? No. You need to provide sources for your human fact-checks. Don’t use an AI system to fact check if that system doesn’t include links to where that information came from.

Generally, I look at AI in this respect the same way I look at Wikipedia. Wikipedia can be very useful. But it should never be your only source. AI shouldn’t be your only fact-checker. You need to fact-check that chatbot when it’s done. So depending on what you are trying to accomplish, you might not be saving any time. (And then you might save enough time to make it worthwhile.)

Factor all of those things in, then make an informed decision about how to approach a specific fact-check.