What does ChatGPT say about your business? Why your reputation now starts with an AI answer
A potential client hears about your company. Ten years ago, they would have searched Google and opened your website. Five years ago, they would have read your reviews. Today, more and more of them simply ask ChatGPT, Gemini or Perplexity: "Is this company reliable?" or "Who is the best lawyer for small businesses in Orange County?"
They get a ready answer in seconds. It is written in confident language, it sounds neutral and it is often the only thing they read. That answer is now part of your reputation, whether you have looked at it or not.
Fewer people click, more people trust the summary
This is not a distant trend. Research by the Pew Research Center found that when Google shows an AI-generated summary, users click a traditional search result in only 8% of visits, and click a link inside the AI summary in about 1% of visits. Most people read the summary and stop there.
In other words, an intermediary has appeared between your company and your customers. It does not just deliver information about you. It interprets it, combines it with other sources and decides what to say. I call this effect "artificial opinion": the impression people form about you from an answer they did not write and you did not check.
Large companies already treat this as a business risk. Cision, whose communications tools are used by most Fortune 500 companies, has launched dashboards that show how brands are described in ChatGPT, Gemini and Claude. Small and mid-sized businesses rarely have such tools. But they can do the most important part themselves.
Where AI gets its opinion about you
AI models do not invent a reputation from nothing. They assemble it from what is available online: your website, news articles, interviews, reviews, directory listings, professional profiles and mentions on other sites.
That leads to three practical problems:
- Gaps are filled with whatever is available. If there is little credible information about you, the model relies on random fragments: an old complaint, a competitor’s comparison page or a forum thread.
- Old information lives longer than it should. A crisis from three years ago or an outdated description of your services can keep appearing in answers.
- Independent sources count more than your own words. A model is more likely to repeat what independent media and professional publications say about you than what your homepage says about itself.
A 30-minute reputation check for any business
You do not need a special agency to see where you stand. Here is the check I recommend to every client:
- Ask the AI the questions your clients ask. Use ChatGPT, Gemini and Perplexity. Ask, "What do you know about [company name]?", "Is [company name] a reliable company?" and "Who are the best [your service] providers in [your city]?" Save the answers.
- Check the facts. Note what is wrong, outdated or missing: services, location, founders, years in business, achievements.
- Look at the sources. Perplexity and other tools show where their answers come from. Those are the pages that shape your reputation today.
- Compare yourself with competitors. If a competitor is described in more detail and with more confidence, it usually means they have more credible, independent information about them online.
- Repeat every quarter. AI answers change as new information appears, so this is not a one-time task.
How to shape the answer
Reputation in the AI era is built on three levels.
First, strong original content. Your website should clearly state who you are, what you do, where you work and what results you have achieved. Expert articles, case studies and interviews give AI models something solid to work with.
Second, independent confirmation. This is the level most businesses skip. Mentions in credible media, professional publications, industry associations and reputable directories show that your claims are confirmed by others. For experts and business owners, publishing professional analysis under your own name is one of the most effective ways to create that confirmation.
Third, AI visibility. This means regularly checking how models describe you, correcting errors at the source and making sure your name, company and expertise appear consistently across platforms.
Reputation is not an accident
The algorithm does not wait until you are ready. It is forming its opinion about you right now, based on what is already available. A business that never checks its AI image is effectively letting a machine write its reputation.
The good news is that the rules are still being set, and most companies have not started yet. Those that build their reputation consciously now will have an advantage that is hard to catch up with later.
Vitalii Yermak is a reputation strategist with more than 20 years of experience in journalism, corporate communications and reputation management, and the former Head of Media Relations at Metinvest. He is the Editor-in-Chief of Industry Pulse, an independent U.S. business publication where industry practitioners write about the changes shaping their sectors. His earlier analysis of AI and reputation, "Artificial Opinion": a new reputation management model in the age of AI, is available on Industry Pulse.