When verifying information is already cognitively difficult—whether from executive dysfunction, processing differences, or attention challenges—relying on AI for facts becomes genuinely risky because you can't easily catch when it's wrong. The solution isn't avoiding AI; it's building external verification systems: trusted sources you pre-identify, another person who fact-checks, or structured tools that cross-reference automatically.
Here's something that trips up new AI users: Sometimes AI will confidently tell you something completely false. It won't sound unsure or hesitant. It will state it as fact. This is called a "hallucination," and it's one of the most important things to understand about AI limits.
Think of it like this: You meet someone who talks confidently about everything, but sometimes makes things up without realizing it. They're not trying to deceive you—they're just confidently filling in gaps. That's what AI does. It's trained to produce coherent, confident-sounding text. Sometimes that text is accurate. Sometimes it's completely made up.
AI doesn't "know" things the way you know things. It's pattern-matching. It sees that certain words go together and produces text that sounds right. If it doesn't know something, it sometimes invents plausible-sounding details instead of saying "I don't know." This is especially bad because the made-up details often sound incredibly real.
The answer is not to stop using AI—it's to verify critical information. If you're using AI to brainstorm or think through a problem, hallucinations don't matter. But if you're using it to make decisions or provide information to others, verify anything that could be wrong.
Simple rule: If the answer is important, fact-check it. Google the claim. Check the source. This takes 30 seconds and saves you from disasters.
Also, ask AI to cite sources. "Where did you get this information?" If it can't tell you or makes something up, that's a warning sign.
Using AI to brainstorm marketing ideas? Hallucinations are fine—you're not taking them as fact, just as inspiration. Using AI to draft a document you'll edit? Fine. Using it to answer factual questions you'll rely on? That's when you verify.
Try this: Ask Claude or ChatGPT a specific factual question about something you know about (maybe a famous person's birth year or a historical date). Then verify the answer with a Google search. You'll see whether it got it right and start understanding your particular AI's hallucination patterns.
Peri can explain this concept, give practical examples, help you decide whether it applies to your situation, or recommend a journey if appropriate.
Explore related journeys or tell Peri what you're working through.