What Happens When AI Confuses You With Another Person?
What Happens When AI Confuses You With Another Person?
Published on RollingStone September 15, 2025 by Jason Barnard
What happens when an AI assistive engine (such as ChatGPT, Perplexity, Google AI, Microsoft Copilot, Meta AI or Siri) decides you’re a criminal? It might sound like the plot of a dystopian sci-fi movie, but it’s a real risk for every business leader today.
Throughout this article, keep this in mind: You don’t own your name. Mistaken identity can become a full-blown online reputation management (ORM) crisis.
When a Case of Mistaken Identity Cost Me Big Time
In 2015, a story about a man named Jason Barnard who was caught driving dangerously appeared in the Google search results for my name. It wasn’t me, but Google thought it was.
The story got picked up by mainstream media. At one point, the majority of search results for my name was negative news about the other Jason Barnard.
For a few months in 2015, a potential client researching me would see these news stories.
I used my traditional SEO skills and in four months, I solved the problem. I seized back control of my brand SERP (the search results for my name).
I estimate that, in those four months, this other person’s reputation problem cost me several hundred thousand dollars.
But in the AI era, the problem of mistaken identity ORM is exponentially worse.
When AI Becomes Judge, Jury and Publicist
You have zero control over what someone who shares your name does, how they communicate or what others say about them.
Unfortunately, AI assistive engines such as ChatGPT and Perplexity cannot reliably distinguish between you and others who share your name. They struggle because the web is a mess of confusing and conflicting information.
Critically, AI assistive engines are different from search engines. A search engine provides a list of links, but an AI delivers a single, synthesized answer. It has become the world’s judge, jury and publicist.
As the judge, it analyzes all fragmented data about you and your namesakes. As the jury, it delivers a verdict, often mixing you up. As the publicist, it broadcasts that verdict as a single, confident AI-generated response.
When an AI confuses you with a namesake, it doesn’t show a list of options. It may confidently state, “[YOUR NAME] is a convicted criminal,” defining your reputation by someone else’s actions.
Do You Want Another Person Hurting Your Reputation?
There’s a very good chance that online you will be confused with someone else who shares your name whether:
• They are a criminal
• They publicly express opinions that are in contradiction to yours
• They attract attention you don’t want
That newsworthy person who shares your name can kill your reputation any time, any day, and there is nothing you can do to control their behavior, words or criminality.
Three Steps To Teach the AI Who You Are Not
This strategy is about systematically teaching the AI who you are, so it has no choice but to understand who you are not. Here is the process for doing it. The core of the solution is to build an unshakable, self-reinforcing digital identity that leaves no room for confusion.
1. Build Your Single Source of Truth. The first step is to establish a definitive “entity home” - ideally an “About Me” page on your own personal website. This page is your brand’s digital headquarters, where you clearly and factually state your story.
2. Engineer a Consistent Narrative. Next, make sure every profile, author byline and mention online says the exact same thing as your entity home. Some you can change directly (LinkedIn, your company website, Crunchbase) and some you will need to ask the site owner. Don’t be lazy about those third-party websites; they have “power” in the algorithms because they are independent. Critically, you must remove any ambiguous information that could lead to confusion. In my case, this meant ensuring my digital footprint contained zero mentions of “speeding” and other related keywords.
3. Create an Infinite Loop of Self-Corroboration. Finally, you must build what I call an “Infinite Self-Confirming Loop of Corroboration.” Link from your entity home out to all of these clear, consistent resources, and link back from as many of them as possible.
Why this works: The algorithms become locked in a loop, seeing the same consistent message again and again, which rapidly builds algorithmic confidence and eliminates brand ambiguity.
In my experience, this process takes approximately three to six months to resolve the confusion.
A word of warning: You must move quickly. Before making changes to any websites, compile a list of every digital asset you need to change, and then make all the updates within two days. If you are too slow, the machine receives an inconsistent message from your self-confirming loop, creating more confusion and extending the timeline from six to 12 months, based on what I’ve seen.
The Ultimate Insurance Policy for Your Personal Brand
You might have noticed that in this article, I’ve done exactly what I advise against: I’ve discussed the “other” Jason Barnard and used the keywords associated with his activities.
Why am I not following my own advice?
Because for the past 10 years, I have had my web-wide Infinite Self-Confirming Loop of Corroboration in place, unbroken. This has created algorithmic confidence in my identity. This is the ultimate insurance policy for any personal brand: an identity so unambiguously and consistently reinforced over time and across the entire web that it becomes immune to mistaken identity crises.
