Themes discussed in this episode

• Why blind trust in AI is the real career risk
• How AI hallucinations can lead to legal, financial, and reputational consequences
• The danger of using AI for delegated thinking instead of supported thinking
• Why long AI chat threads may increase inaccurate outputs
• Practical guardrails for using AI responsibly at work

Episode Highlights

Time-stamped inflection points from the show

00:12 – Lori opens with a real courtroom example of hallucinated case law and explains how AI mistakes can become career crushers.

01:36 – A firsthand story of AI overstating mailing list data by 100x — and why confidence doesn’t equal correctness.

02:15 – The Reddit story: executives making business decisions based on unchecked AI analytics.

03:16 – Why long AI conversations may increase hallucinations and how to properly verify sources.

04:38 – Practical AI guardrails: verify facts, cross-check numbers, restart long threads, and triple-check high-risk data.

05:18 – AI should accelerate your thinking, not replace it — and why trusting AI more than your own brain can cost you your job.

Top Quotes

03:10 – “Don’t get caught by AI hallucinations. You don’t want it to happen.”

04:25 – “You don’t want to use AI for delegated thinking. You want to use it to support your work and your thinking. It’s a draft partner, not a decision maker.”

05:46 – “Don’t be the person who gets fined or fired because you trusted AI more than your own brain.”

---

Transcript

00:12

Hello, I’m Lori Jo Vest. Thank you so much for joining me for this episode of Work Mom Says Don’t Be an Idiot, and I’m here to help you play the emotional contact sport of business so that you can experience less drama and more success. So today we are talking about AI and how it can be a workplace trap. Be careful. Just recently, there was an attorney that submitted a brief to a judge in a courtroom that had hallucinated case filings that it had referenced. So it actually went out there and not only looked for case filings, but it made some up, and it went all the way to the judge, and those cases that were referenced didn’t even exist. So you can see how that could be a career crusher, right?

01:02

You can be a smart person with really good intentions and still get it totally wrong if you’re using AI, because the problem isn’t AI. The problem is blind trust of AI. It’s not the villain. Overconfidence is the villain. So honestly, I’ve been pinged by it myself, where I uploaded a spreadsheet of information about a mailing list, and it gave me all this great analysis of what was in it, but it was completely wrong.

01:36

One example — I asked about a particular type of name, you know, a particular category. Did we have a lot of people in this industry on our mailing list? And AI said, yes, you have 4,000 of those people. We didn’t have 4,000 of those people. We had 40 of those people. It was working from my data that I uploaded, and it still completely gave me the wrong number. So just because AI sounds confident, just because it references a source, does not mean it’s correct. You’ve met people like that. AI is a person like that. So be careful.

02:15

So why is AI a career risk right now? Well, because a lot of people have way too much confidence in it, and they’re using it for things like projections and analytics. My husband was just reading something on Reddit over the weekend that he mentioned to me, and I was just like, oh my God. And it was a marketing department that had taken their analytics and uploaded it to an AI tool and had it analyze it for the last, you know, three to six months, and that’s the data that they gave to their C-level leaders who made business decisions based on that data. And three to six months later, when they actually looked at the data and compared it to the source, it was completely incorrect, and that person that made the decision to submit that AI data without checking it is in an incredible hot seat. I cannot imagine what that would feel like. Don’t get caught by AI hallucinations. You don’t want it to happen.

03:16

A couple points. Some of the things that happen that they’ve noticed is that if you are using one long chat on the same topic, you may feel like the more you use this same chat for information on a particular topic, the more informed it will be. In fact, the polar opposite is true. They say that hallucinated data is more likely to occur in long AI chats. So something to think about. What I tell people to do, and what we do at the agency that I run, is we actually ask for sources. We ask, is this correct? Is this true? Show us the sources. And then we go look at the sources. Because one of the things that happens with AI is it will look at a source, grab a couple things, and present data with that source. But when you go look at the actual source, the article may not say that at all. So if you’re using data from an AI chat, ask for the source and go look and double-check it, because you don’t want to use AI for delegated thinking. You want to use it to support your work and your thinking. It’s a draft partner, not a decision maker. So be extremely careful.

04:38

And here’s how to use AI without looking like an idiot. Verify anything factual. Always, always ask for sources. Cross-check your numbers. When long threads are happening and you realize you’ve gone, you know, a couple scrolls down — it’s a long conversation with chat — go back, start a new one, and that will help ensure that the information is more accurate. But again, still double-check it. If it has anything to do with money, law, health, your reputation, or data for decision making, double-check it. Matter of fact, triple-check it, because it’s going to get a lot of people in trouble, because we just have too much confidence in it right now.

05:18

So what I would tell you is make sure that the tools you’re using accelerate your thinking — don’t replace it. Don’t be lazy. Honestly, I can say that I’ve done it. I’ve gone, oh yeah, I’m sure that’s accurate, and sent it along. Don’t learn the hard way. One of the things that I always say as Work Mom is I learned the hard way so you don’t have to. Always make sure that you’re double-checking the information, and don’t be the person who gets fined or fired because you trusted AI more than your own brain. Your own brain is always going to give you better information, and you can affirm what you find in AI. So what I would say is do me a favor. Share this with your team. Share this with people around you, and talk about AI guardrails at work. Talk about how you can build verification into the work that you do so that you don’t ever get stuck with an AI hallucination causing big problems for you.

06:17

So that’s what I’ve got for you today. If you are watching this on YouTube, please do me a favor — subscribe, like, give me a comment. I’d love to hear what you want me to talk about. We have a website at workmomsays.com where you can go and send me a message and let me know if you have a person that you’d like me to invite onto the show, or if you personally think you’d make a good guest. If you have a topic or a question you’d like me to answer, that would be great. I’d love to hear from you. I’m also an open networker on LinkedIn at Lori Jo Vest. I would love to hear from you there as well. So that’s what I’ve got for you. I will be back soon with some more helpful information to help you play the emotional contact sport of business. Take care.

---

Who is our ideal listener?

This podcast is for young professionals who want to learn to play the emotional context sport of business and experience less drama and more success.

How can you be more logical and less emotional? Be strategic, and Work Mom Says can help you.

“I tell people to back up, put down the magnifying glass, and look at the big picture when you’re responding to something,” said Lori Jo Vest, Work Mom. “In doing this, you will understand that what’s really upsetting you right now will be something you don’t even remember next week.”

What value can people get from listening to this podcast?

Listening to Work Mom Says can help you grow your mood management skills, grow your ability to reframe situations, and look at things from a strategic point of view. This makes it easier to go into a work situation and get the most positive results.

On Work Mom Says, we also offer tips and tricks for creating connected positive relationships that last over time. People will want you on the team if you can create connected positive relationships and work environments. You become an asset, and you will be more successful when you’re an asset.

“I also like to talk about developing traits like optimism, persistence, tenacity, stick-to-itiveness, sticking with things, and approaching every project with a curious mind instead of a fearful mind,” said Lori Jo Vest, Work Mom

                                                                                                     

Why do I do this? A few more words from Work Mom

I do this because I naturally fell into the Work Mom role when I worked in the ad agency business and had so much fun with it. I also realized I had made just about every mistake there was to make. I don’t hold myself as a stellar example of truth and how you should be. I hold myself out there as someone who has been bruised, battered, and beaten up and learned some important lessons. I’d love to share these lessons with young people, so they don’t have to make those same mistakes or be the idiot I was.

I also want to help young professionals realize that many things our culture prioritizes aren’t really important. We talk a lot about what should be important and how to present your best face at the office so that you can succeed.

I’ve learned so much throughout my career, and it’s gratifying to share that with young professionals and help them avoid some of those mistakes and get to that success sooner.