I recently delved into the intriguing world of Answer Engine Insights and discovered a groundbreaking update: eight distinct citation categories. These categories reveal the true sources of AI visibility.
This update provides a fresh perspective, as it’s backed by insights derived from analyzing an impressive 85 million citations across ChatGPT, Gemini, and various AI Overviews. Now, isn’t that fascinating?
If you’ve ever wondered about the mechanics behind AI answers, this exploration into citation categories might just provide the clarity you’ve been seeking.
Every week, 900 million users interact with platforms like ChatGPT, yet my brand is often absent from their discussions. I wanted to understand why, and more importantly, how I could change that.
After digging into the reasons, I discovered several factors that influence brand visibility within ChatGPT’s responses. It’s not just about having a strong brand, but also understanding how to optimize for this new kind of search engine.
To make my brand appear in ChatGPT talks, I need to focus on implementing strategic optimization techniques, specifically tailored to AI-powered platforms. I’ll share these insights, so we all can benefit from increased visibility in this rapidly evolving space.
I’ve been closely following OpenAI’s journey as they pause ChatGPT ads to focus entirely on optimizing the user experience. It’s a daring decision, and I see it as a strategic move to challenge Google’s Gemini’s dominance in the AI landscape without distractions.
For years, as the forefront of AI innovation with ChatGPT, OpenAI seemed unbeatable, especially with their partnership with Microsoft. However, tables have turned, and the competition is heating up with Google’s Gemini gaining ground and even surpassing in vital areas.
When OpenAI CEO Sam Altman announced an internal “code red,” I realized this was a wake-up call to prioritize ChatGPT’s quality over everything else. This pause meant putting their advertising plans on hold, not forgoing them entirely.
It’s fascinating to me how OpenAI is handling this situation. The focus is on fixing fundamental issues related to speed, reliability, and reasoning to retain their user base. Despite the pause, advertisements are still part of the long-term strategy.
This leads me to wonder: what steps is OpenAI taking to catch up, and what does this delay mean for the future of AI advertising? Understanding these aspects is crucial for predicting OpenAI’s path forward.
Examining the performance shift, I see that OpenAI and Microsoft weren’t slowing down. Instead, Google’s investment in infrastructure paid off, exposing weaknesses in OpenAI’s alliance. The key lies in model architecture, as Google’s Gemini 3 is built as a “native multimodal” model, unlike ChatGPT’s combined approach, which feels less cohesive over time.
Google’s advantage of owning the technology that powers Gemini offers them unbeatable optimization and cost control. OpenAI faces challenges with their reliance on costly Nvidia GPU integrations.
This lack of an all-encompassing ecosystem is contributing to the shift in user sentiment towards Google. Users experience Gemini as a unified assistant embedded into their daily work routine, in contrast to the slightly disjointed feel of Microsoft’s Copilot.
I find it telling that Gemini now outperforms ChatGPT in benchmarks for reasoning and speed, highlighting the effectiveness of Google’s integrated machine approach over the Microsoft-OpenAI alliance.
Considering how ChatGPT and Gemini tackle the same problems differently, it’s intriguing to see Gemini’s practical approach compared to ChatGPT’s fact-providing nature. Gemini offers real-time solutions by integrating with Google Maps and Workspace, crafting an end-to-end experience that truly solves user problems.
The “code red” response from OpenAI highlights their understanding that without a solid foundation, introducing new features is futile. This realization is driving the development of GPT-5.2, aimed at closing the gap with Gemini in complex reasoning and coding.
OpenAI is focused on stopping hallucinations, improving speed, and making the interaction feel intuitive and personal again. They aim to move from a passive chatbot to a reliable executor of complex tasks, an area where Google currently leads.
For Microsoft, the challenge is to unify the Copilot experience, solving data silo issues. They need to leverage Office 365 data more effectively, akin to Google’s personalization using user data.
The pause on ad deployment serves as a significant indicator of OpenAI’s strategic priorities. Introducing paid ads amid current challenges would risk user loss, and OpenAI understands the necessity of retention before revenue.
OpenAI recognizes that to introduce advertising successfully in the future, the product must stabilize against Gemini’s advancements. When trust is restored, only then can monetization through ads be pursued.
The delay allows OpenAI to craft ad formats that are integrated and contextually relevant, ensuring they enhance rather than disrupt user experience. I believe that properly executed ads will become an essential revenue stream.
Overall, pausing ChatGPT ads reflects a necessary strategy to refine its core capabilities and challenge Google’s dominance effectively. In doing so, OpenAI hopes to reclaim its position and eventually introduce ads that align seamlessly with user expectations.
I recently delved into OpenAI’s intriguing move towards integrating ads within their AI responses, which could potentially transform digital marketing by providing advertisers with a highly contextual channel.
OpenAI is laying the foundation for developing an advertising model, which could signify a shift from solely relying on subscriptions and enterprise agreements for revenue generation.
Unpacking the Strategy. As reported by The Information, OpenAI is in the early stages of discussing ad formats and partnerships, potentially placing ads within AI-generated responses. These discussions are preliminary, but it’s clear that ads are becoming an integral part of OpenAI’s long-term financial strategy.
The Implications. This exploration into ads embedded in AI responses offers a unique opportunity to reach users right as they seek information. It positions OpenAI to compete with industry giants like Google and Meta, while also raising questions about user trust and engagement. Early adopters might gain a competitive edge since the dynamics differ from traditional digital ads, marking a new era in advertising.
Navigating User Experience. OpenAI seems to be cautiously approaching this initiative, focusing on maintaining a seamless user experience and not watering down trust in their AI models. Initially, any ad implementation will likely be carefully curated and contextually relevant to enhance, rather than disrupt, user interactions.
Considering the Broader Impact. Given the escalating costs of infrastructure and the increasing demand for revenue growth, integrating ads could be pivotal for OpenAI. This becomes especially relevant as generative AI continues to redefine how users search for information and discover products.
Future Developments. As these ad plans evolve from internal planning to public trials, a critical area to watch will be how transparently they are implemented and whether users will embrace ads within AI-driven results.
In Conclusion. OpenAI isn’t hurrying to deploy ads on the market, but the groundwork is being set. Their eventual full-scale deployment could reshape not only AI tools but also the digital advertising landscape as we know it.
Back in June 2025, I noticed an interesting infographic circulating widely. It highlighted the most-cited websites by AI models, according to a comprehensive SEMrush study of over 150,000 citations. Naturally, seeing Reddit at the top sparked a buzz among marketers, who began to believe that featuring on Reddit was key to their GEO strategy.
But, here’s the catch: most of these citations were research-oriented, not necessarily geared towards buying intents. For instance, Reddit was frequently mentioned in queries like “Where in Europe should I take a family vacation?” but not so much in “What are the best web design firms?”
This led to some misguided assumptions. Although Reddit does play a role in AI models like ChatGPT, Perplexity, and Gemini, it represents only about 11% of their commercial recommendation algorithms. So, placing too much emphasis on Reddit won’t really boost your product or service visibility in these AI models.
The Most Cited Websites by AI Models for Buying-Intent Queries
Our research team dug deeper into this in October 2025 and later updated it in December 2025. We conducted 36,127 buying-intent queries on ChatGPT and tallied the top-cited websites. Our “buying-intent query” was defined on a scale measuring how close a query was to a purchase decision. A simplified version of this scale is captured in the infographic below:
A query scoring higher than 1.35 was marked as “buying-intent.”
Top Website Types Cited by AI Models for Buying-Intent Queries
We meticulously categorized the types of websites AI models prefer for such queries. Understanding the types helped us unravel which channels are more effective at GEO—essential in influencing AI chatbots to nominate certain companies.
Top Website Types Cited by AI Models for Buying-Intent Queries
#
Website Type
Description
# of Citations
1
Product Recommendation Media
“Best of” and “Top 10” review sites largely monetized via affiliate links (e.g., Wirecutter, Tom’s Guide, TechRadar).
7,642
2
Consumer Review Platforms
User-generated review aggregators like Trustpilot, BBB, and Google Reviews.
5,983
3
Traditional Media
Established publishers including product roundups or consumer coverage (Forbes, NYT, Wired).
4,581
4
New Media
Digital-native outlets that frequently review or endorse products (TechCrunch, The Verge).
3,826
5
YouTube / Video Review Channels
Video-based reviews and product comparisons often transcribed or summarized by AI models.
Regulatory or authoritative institutional content (FDA.gov, FTC.gov).
159
20
Standards & Certification Bodies
Official verification or compliance organizations (UL, ISO, Energy Star).
119
The major takeaway from parsing this data? Websites offering list-based product recommendations feature heavily in AI rankings. Being listed on these commercial publications can greatly enhance a product’s visibility in AI recommendations like ChatGPT, Perplexity, and others.
Industry Breakdown: Websites Most Cited by AI Models for Buying-Intent Queries
Next, our analysis focused on the top three websites each industry traditionally relies on. This provided a glimpse into which platforms AI models commonly heed within specific verticals, giving GEO marketers a decisive edge in targeting their media placement efforts.
Top Websites Cited by AI Models in Buying-Intent Queries, by Industry
Industry
Top-Cited Websites by AI for Buying-Intent Queries
eCommerce
Wirecutter, Forbes, Tom’s Guide
Managed Services
Clutch, G2, UpCity
Healthcare
Forbes Health, Verywell Health, Medical News Today
A key insight here is the fragmented nature of website citations. Trade journals contributed more than the top three sites in any category. As with website types, the predominance of review sites and product recommendation platforms was notable.
Questions, Media Inquiries, or Other Requests
Curious about our study? Have a media request, or want a PDF copy? Reach out to us here.
On December 11th, I was excited to learn about OpenAI’s announcement of GPT-5.2, their most advanced frontier model yet. Knowing how transformative AI can be, I was thrilled to see that Profound is now tracking GPT-5.2 responses in ChatGPT. This upgrade is integrated across our entire product suite, including Answer Engine Insights, Prompt Volumes, and Agent Analytics.
Starting today, every ChatGPT response on my dashboards reflects this cutting-edge model. It’s exhilarating to think about the enhanced capabilities and strategic insights that I can now access thanks to this update.
I’m thrilled to share that OpenAI has announced its latest advanced model, GPT-5.2, on December 11th. This groundbreaking development offers incredible capabilities, and starting December 14th, I’m excited to let you know that Profound is now tracking GPT-5.2 responses across all our products, enhancing your experience with ChatGPT.
From Answer Engine Insights to Prompt Volumes and Agent Analytics, every ChatGPT response in your dashboard now reflects the prowess of GPT-5.2. This update signifies a leap forward in the responsiveness and accuracy of interactions, providing you with richer insights and more refined outputs.
Let’s dive into this new era of AI on Profound, where the boundaries of what’s possible continue to expand daily. Whether you’re using it for research, business insights, or simply exploring the model’s capabilities, you’ll notice a distinct enhancement in all ChatGPT interactions.
Stay tuned as we continue to innovate and incorporate these advancements into your everyday tools. I can’t wait for you to experience the phenomenal benefits that GPT-5.2 brings to the table.
I recently came across a discussion where OpenAI needed to reassure us, their paying users, about some rather curious developments with ChatGPT.
To my surprise, images emerged of what seemed to be ads from big names like Target and Peloton. Understandably, this concern had me quite intrigued because OpenAI insisted these prompts weren’t actually ads.
What Happened. Many of us who use ChatGPT regularly shared screenshots that displayed prompts resembling promotional ads. One notable instance was a message encouraging shopping for home and groceries with Target.
OpenAI clarified that these prompts were merely recommendations for apps built on their platform, emphasizing that there was “no financial component” involved.
Despite this, the presence of brand logos and calls to action led us to believe otherwise. It naturally felt like these were ads.
Not surprisingly, many of us weren’t thrilled about this. One user responded with palpable frustration, saying, “Bruhhh… Don’t insult your paying users.”
What OpenAI is Saying. OpenAI’s management was quick to address our concerns:
“No ads.” Nick Turley, the head of ChatGPT, insisted that there are currently “no live tests for ads.” He suggested that the screenshots we saw were “either not real or not ads.”
However, Mark Chen, the chief research officer, admitted that OpenAI “fell short” and any content that could be perceived as an ad should be “handled with care.”
Chen also mentioned the disabling of these app suggestions while they refine the model’s accuracy and build better controls for users to manage them.
Ads on Hold. Despite all the chatter about potential ads on ChatGPT, OpenAI has put a pause on their advertising ambitions. CEO Sam Altman issued a companywide “code red” to enhance ChatGPT’s quality, leading to a delay in several initiatives, including those related to ads.
Why We Care. Initially, it seemed like we were about to see the dawn of ChatGPT ads, but that isn’t the case. For now, we eagerly await the end of ChatGPT’s “code red” phase and look forward to a more refined and improved product experience.
More Coverage. Discover further discussions on this issue at Techmeme.
Today, I noticed that ChatGPT, the popular AI service from OpenAI, was unresponsive for many of us trying to interact with it. Attempts to ask questions yielded no results, leaving many in a lurch.
After checking OpenAI’s status page, I found a confirmation of the issue. OpenAI has stated, “We’re currently experiencing issues.”
They further explained, “We have identified that users are experiencing elevated errors for the impacted services. We are working on implementing a mitigation.” This means they’re actively working to resolve the problem.
What it looks like. For many users, attempting to use ChatGPT results in seeing just a black dot instead of an answer, which is quite frustrating.
Offline for many. Visiting Downdetector, I saw that thousands of users have reported ChatGPT as offline, indicating a widespread issue.
The issue first surfaced around 2:30 PM ET, and even after 30 minutes, it remains unresolved.
Why we care. Whether I’m using ChatGPT for email, marketing, SEO, or PPC tasks, this outage disrupts the workflow significantly.
I’m confident that OpenAI is making every effort to resolve these issues promptly, and I hope to see it operational soon.
