Key points
- ChatGPT reached 900 million weekly active users in February 2026, according to The Next Web.
- Google’s Gemini app passed 750 million monthly users by the end of 2025.
- AI tools generate answers based on training data snapshots and real-time web searches.
The landscape of online reputation management has shifted as artificial intelligence platforms replace traditional search engine result pages with concise, one-paragraph summaries. According to a report by The Next Web, these summaries are fast becoming the first impression for users, eliminating the ability to push negative content to later pages of search results. This change forces a reevaluation of how personal and professional reputations are monitored and managed in the digital sphere.
How it unfolded
- ChatGPT achieved 900 million weekly active users in February 2026, establishing a massive audience for AI-generated answers.
- Google’s Gemini application surpassed 750 million monthly users by the end of 2025, further expanding the reach of AI search.
- AI tools rely on frozen training snapshots and real-time web searches to construct responses about individuals and entities.
Who is affected
Organizations and public figures are affected because AI tools do not allow users to edit their profiles directly. Every answer generated by these platforms is derived from two distinct sources: historical training data and real-time web searches. The Next Web reports that models learn from a frozen snapshot of the web, meaning outdated information such as old lawsuits or harsh reviews may persist in answers even after the underlying facts have changed or been resolved. This creates a persistent risk for reputation damage that is difficult to correct through traditional search engine optimization techniques.
The specific sources trusted by each platform vary significantly, leading to inconsistent information across different tools. ChatGPT leans heavily on encyclopedias and news sites, with Wikipedia accounting for 47.9% of its top-10 cited sources. In contrast, Perplexity trusts online communities most, with Reddit comprising 46.7% of its top-10 cited sources. Google Gemini and AI Overviews draw from a mix of social, video, and professional sites, with Reddit, YouTube, and LinkedIn appearing in their top-10 citations. These preferences are not fixed and can change quickly without warning, adding further complexity to reputation management efforts.
What happens next
The divergence in source priorities means that a single question about an individual or organization can yield three different answers depending on which platform is used. According to The Next Web, the sites a platform trusts can shift rapidly, making long-term reputation strategies volatile. Anthony Will, co-founder and CEO of Reputation Resolutions, states that Google and AI search results are the first impression to the world. He emphasizes that AI tools repeat what they find online rather than inventing reputations, suggesting that fixing the underlying web sources fixes the answer.
This dynamic separates firms that specialize in AI reputation management from those focused solely on traditional search results. The work involves identifying and correcting the primary sources that AI models rely on, whether they are Wikipedia pages, Reddit threads, or news articles. As the user bases for ChatGPT and Gemini continue to grow, the pressure to manage these diverse data points will likely intensify for system administrators and business leaders alike.
What to do and how to stay safe: ChatGPT
- Audit the presence of your organization or personal brand on high-trust sources like Wikipedia, Reddit, and LinkedIn, as these are heavily cited by AI tools.
- Monitor multiple AI platforms (ChatGPT, Perplexity, Gemini) for their respective answers about your entity, noting discrepancies in cited sources.
- Correct outdated or inaccurate information at the source level on websites and social platforms, as AI tools repeat existing web content.
- Stay alert for changes in platform behavior, as the sources AI tools trust can shift quickly and without prior notice.
Step-by-step guide: Injection attacks examples: how prompt leaks break AI systems
General security guidance from the Circuit Observer newsroom. It is not confirmed advice from the organisations named in this story.
Frequently asked questions
Do AI tools invent negative information about people or companies?
According to The Next Web, AI tools do not invent reputations but repeat information they find on the web, drawing from training data and real-time searches.
Which sources does ChatGPT trust most for its answers?
ChatGPT leans on encyclopedias and news sites, with Wikipedia making up 47.9% of its top-10 cited sources.
How does Perplexity differ from ChatGPT in sourcing information?
Perplexity searches for every answer and shows numbered citations, trusting online communities most, with Reddit at 46.7% of its top-10 cited sources.
