Author name: aftabkhannewemail@gmail.com

The New Era of Monetization: ChatGPT Ads Are Here For the better part of two years, ChatGPT has been the crown jewel of the generative AI revolution, offering a clean, conversational interface that felt remarkably different from the cluttered, ad-heavy experience of modern search engines. However, that “honeymoon phase” for free-tier users appears to be coming to an end. Recent reports and user data suggest that OpenAI is aggressively ramping up its advertising efforts, integrating sponsored content directly into the flow of AI conversations. As the company expands this rollout from a pilot phase in the United States to international markets, the reality is clear: ChatGPT ads are showing up—and they are showing up a lot. This shift represents a fundamental change in how users interact with artificial intelligence. What was once a pure information retrieval and creative tool is now transforming into a powerful marketing platform. For OpenAI, this is a necessary step toward sustainability and profitability. For users and digital marketers, it is the beginning of a new chapter in digital advertising that combines the precision of search intent with the nuance of conversational context. The Data Behind the Rollout: Frequency and Reach The transition from a completely ad-free experience to one integrated with sponsored links has been swift. Testing conducted on the ChatGPT mobile app reveals just how pervasive these ads have become. In a controlled test involving 500 unique questions, it was discovered that roughly one in five questions—or 20% of interactions—triggered a sponsored ad at the bottom of the AI’s response. These ads typically manifest as “website link buttons.” They are positioned discreetly but prominently enough to catch the eye once the AI has finished generating its text output. Notably, these ads are currently restricted to the free-tier user base. Users subscribed to ChatGPT Plus, Team, or Enterprise versions have not reported seeing these placements, suggesting that OpenAI is maintaining a clear value proposition for its paying customers: pay for the service or be the product for advertisers. Initially launched as a pilot program in the United States, the advertising infrastructure is now being scaled globally. Following the U.S. testing phase, OpenAI has begun expanding these ad placements to users in Canada, Australia, and New Zealand. This rapid expansion suggests that the early results from the U.S. market met OpenAI’s internal benchmarks for performance and user retention. How ChatGPT Ad Targeting Works One of the most significant advantages OpenAI holds over traditional search advertising is the depth of context it possesses. Unlike a search engine that relies primarily on a single query and perhaps some browser history, ChatGPT has access to the entire thread of a conversation and a “memory” of past interactions. OpenAI has clarified that ad targeting is based on three primary pillars: 1. Topic of the Current Question The most immediate factor is the subject matter being discussed. If a user asks for a recipe, they might see an ad for a grocery delivery service or a specific cookware brand. This is high-intent targeting that mirrors the “contextual advertising” we see on blogs, but with much higher relevance because the AI understands the user’s specific problem. 2. Past Chat History Because ChatGPT can reference previous interactions within a thread, the ads can evolve as the conversation progresses. If a user starts by asking about a trip to Europe and later asks about packing tips, the ads may transition from flight bookings to luggage brands. 3. User Memory The “Memory” feature in ChatGPT allows the model to remember specific details about a user over long periods—such as their dietary preferences, their job title, or their hobbies. This data provides a rich profile for advertisers to target users with uncanny accuracy without the advertiser ever seeing the raw conversation data itself. The “Poaching” Dynamic: A New Battlefield for Brands One of the more controversial aspects of the new ChatGPT ad ecosystem is what marketing experts call “brand poaching.” This occurs when a user mentions a specific brand in their prompt, but the ad that appears belongs to a direct competitor. For example, if a user asks ChatGPT to “compare Netflix subscription plans,” the response might be accompanied by an ad for Hulu or Disney+. Similarly, questions about DoorDash might trigger ads for Uber Eats. This tactic is a staple of Google Ads, where brands bid on their competitors’ names to capture “switchers.” Bringing this dynamic to an AI interface feels more intimate and potentially more persuasive, as the user is already in a “consultative” mindset with the AI. For established brands, this means that even if they are the subject of a positive AI response, they are at risk of losing the final click to a competitor who has paid for the link button at the bottom of the chat. This creates a defensive necessity for brands to bid on their own names or ensure they have a presence within the OpenAI ad network. Travel and Tech: The Most Targeted Categories While ads are appearing across a wide variety of topics, certain sectors are being targeted more aggressively than others. Travel, in particular, has emerged as a major category. When users ask for help planning trips or looking for things to do in specific cities, the conversion potential is massive. In one instance, a query about planning a trip to Palm Springs immediately surfaced a Booking.com ad. This wasn’t just a generic link; it was a deep-linked button that automatically initiated a search for hotels in Palm Springs upon being clicked. This level of seamless integration reduces friction for the user and increases the value for the advertiser. Other frequently appearing ad categories include: Productivity Software: Tools for project management and team collaboration. AI and Coding Tools: Promoting specialized AI assistants for developers. Financial Services: Corporate credit cards and accounting software for business-related queries. Consumer Goods: Ranging from dog food to basketball tickets. Streaming Services: Capitalizing on entertainment searches. OpenAI’s Stance on Privacy and Influence The introduction of ads in a tool as

The New Reality of Conversational AI: ChatGPT Ads Take Center Stage For a long time, the promise of generative AI was a clean, uninterrupted interface where users could interact with information without the clutter of traditional search engine results pages. OpenAI, the leader in this space, initially positioned its flagship product, ChatGPT, as a premium tool supported largely by subscriptions. However, the landscape is shifting rapidly. Recent data and user reports indicate that OpenAI has significantly increased its advertising efforts for free-tier users, marking a major turning point in how the platform operates and how it plans to monetize its massive user base. The introduction of ads within ChatGPT is no longer a subtle experiment. It is a full-scale rollout that is currently impacting users across the United States, with plans already in motion to expand to international markets including Canada, Australia, and New Zealand. For the digital marketing industry and the average user alike, this signals the end of the “ad-free” era of conversational AI and the beginning of a new, highly targeted advertising frontier. The Data Behind the Rollout: Frequency and Format Recent investigations into the frequency of these advertisements reveal a high level of integration. In a controlled test involving 500 unique questions posed to the ChatGPT mobile app, the results were striking. Approximately one out of every five questions—or 20% of interactions in a new conversation thread—triggered an advertisement at the bottom of the response. This frequency suggests that OpenAI is not just testing the waters but is actively leaning into an ad-supported model for its non-paying users. The format of these ads is distinct from the banners or pop-ups we typically see on websites. Instead, they appear as “website link buttons” positioned directly beneath the AI’s generated text. These buttons are often accompanied by a brief call to action or a brand name, designed to look like a natural extension of the helpful advice provided by the AI. This seamless integration is part of what makes the ads both effective and, to some critics, potentially intrusive. What Categories Are Dominating the Ad Space? The range of advertisers currently utilizing ChatGPT’s inventory is surprisingly diverse. The test results showed that ads spanned across numerous industries, demonstrating the platform’s broad appeal to various market segments. Some of the most common ad categories included: Travel and Hospitality: This is currently the most active sector. Questions regarding trip planning or specific destinations frequently triggered ads from major players like Booking.com. Software and Productivity: AI coding tools, corporate credit cards, and productivity software ads often appeared when users asked technical or business-related questions. Consumer Goods: Everything from dog food to streaming services and basketball tickets appeared in the results. Financial Services: Ads for corporate financing and fintech tools targeted users engaged in professional or financial queries. The high frequency of travel ads is particularly noteworthy. When a user asked for assistance in planning a trip to a specific location, such as Palm Springs, the AI didn’t just provide a list of things to do; it provided a direct link to a booking engine that had already prepopulated the search parameters for that specific location. This level of utility-driven advertising is a significant evolution from the static ads of the past. Advanced Targeting: Context, Memory, and Intent What sets ChatGPT ads apart from traditional search ads is the depth of the targeting. While Google Search relies heavily on the specific keywords typed into a search bar, OpenAI is leveraging the unique “memory” and conversational context of its platform. According to OpenAI, ad targeting is influenced by three primary factors: 1. The immediate topic of the current conversation. 2. Previous chat history within the same thread. 3. Information stored in the user’s “Memory” profile (if enabled). This means the ads are not just responding to what you asked *now*, but what the AI knows about you from previous interactions. If you have previously discussed an interest in vegan cooking and then ask for a restaurant recommendation, the AI can theoretically surface an ad for a vegan meal delivery service because it “remembers” your preferences. This intent-based targeting is highly valuable to advertisers but raises new questions about how much data is being utilized to serve these placements. The Privacy Question: What Does OpenAI Share? OpenAI has been proactive in addressing potential privacy concerns to maintain consumer trust. The company maintains that the full content of a user’s conversation is not shared directly with advertisers. Instead, the system acts as a middleman—the AI understands the context and then requests a relevant ad from its inventory without exposing the raw transcript to the third-party brand. Furthermore, OpenAI states that the ads do not influence the actual content of the AI’s responses. The “brain” of the AI remains focused on answering the prompt, while the ad system simply attaches a relevant link to the end of the output. The “Poaching” Dynamic: A New Battlefield for Brands One of the most interesting developments in the ChatGPT ad ecosystem is the emergence of “brand poaching.” In the world of traditional Search Engine Marketing (SEM), it is common for brands to bid on the names of their competitors. For example, a food delivery service might bid on a competitor’s name so that their ad appears at the top of the search results when a user looks for that specific rival. This same dynamic has now migrated to ChatGPT. Testing has shown that when a user mentions a specific brand by name—such as asking for the price of a Netflix subscription or the delivery range of DoorDash—the ad button that appears might actually be for a direct competitor. This “poaching” allows smaller or rival brands to intercept a user at the exact moment they are thinking about a specific service. For marketing professionals, this creates a new layer of brand protection necessity. Companies must now consider whether they need to be present on ChatGPT not just to find new customers, but to defend their own brand mentions from

The Rapid Rise of OpenAI’s Advertising Powerhouse In the fast-paced world of digital marketing, few platforms have transitioned from “experimental tool” to “revenue giant” as quickly as ChatGPT. Only six weeks after launching its initial advertising pilot, OpenAI has announced a staggering milestone: the platform has already reached $100 million in annualized ad revenue. This achievement is particularly notable because it has been reached while the advertising program is still in its infancy, operating with significant restrictions on both its user base and its advertiser pool. For months, the industry speculated on how OpenAI would bridge the gap between its massive operational costs and its path to profitability. While the ChatGPT Plus subscription model has been a primary driver of income, the introduction of a sophisticated advertising tier represents a fundamental shift in the company’s business strategy. As OpenAI prepares for a potential IPO and eyes a total revenue target of over $17 billion by 2026, the success of this ad pilot serves as a critical proof of concept for investors and brands alike. The message is clear: ChatGPT is no longer just a productivity assistant; it is becoming a major player in the global digital advertising ecosystem, rivaling the early growth trajectories of search engines and social media giants. Breaking Down the $100 Million Milestone To understand the magnitude of reaching $100 million in annualized revenue in just six weeks, one must look at the specific constraints under which this revenue was generated. Currently, OpenAI is not showing ads to its entire user base. In fact, the ads are only appearing to a fraction of those who use the platform. According to internal data, the revenue was generated from less than 20% of eligible “Free” and “Go” tier users in the United States who interact with the platform daily. Approximately 85% of all Free and Go users are technically eligible to see ads, meaning the current revenue represents only a tiny slice of the platform’s eventual inventory capacity. If OpenAI were to flip the switch and show ads to 100% of its eligible global audience, the revenue figures could easily scale into the billions within a very short timeframe. Currently, the platform hosts more than 600 advertisers. These early participants are part of a managed pilot program, essentially an invite-only phase where OpenAI works closely with brands to ensure the ad placements feel native to the conversational experience. The fact that such a small group of advertisers, targeting a small percentage of users, has already generated $100 million in annualized value suggests that the “intent” behind ChatGPT queries is incredibly valuable to marketers. The April Launch: Self-Serve Access for All While the current pilot is restricted to a curated list of high-spend brands, OpenAI has confirmed that it is on track to launch self-serve advertiser access in April. This is the moment the floodgates will truly open. Self-serve platforms, such as Google Ads and Meta’s Ads Manager, are what allowed digital advertising to democratize. It enables small and medium-sized businesses (SMBs) to bid on keywords and audience segments without needing a direct line to a sales representative at the tech company. When ChatGPT opens this functionality in April, we can expect a massive influx of diverse advertisers, ranging from local services to boutique e-commerce brands. This transition from a managed pilot to a self-serve model typically leads to increased competition for ad placements. For early movers, this creates a “gold rush” window. Historically, those who master a new ad platform during its early self-serve phase—like the early days of Facebook Ads or Amazon Advertising—benefit from significantly lower Cost Per Click (CPC) and higher Return on Ad Spend (ROAS) before the market becomes saturated and prices rise. Strategic Leadership: The Influence of Dave Dugan A key factor in OpenAI’s rapid monetization success is its recent talent acquisition. To lead its ad sales division, OpenAI hired Dave Dugan, a former high-level executive from Meta. At Meta, Dugan was instrumental in scaling one of the world’s most sophisticated and profitable advertising engines. Dugan’s appointment signals that OpenAI is not interested in simply “testing” ads; they are building a professional-grade advertising infrastructure designed to compete with the biggest names in the industry. His background suggests that OpenAI’s ad platform will likely lean into sophisticated targeting, conversion tracking, and perhaps most importantly, “agentic commerce.” Agentic commerce refers to the ability of an AI agent to not just recommend a product, but to help a user complete a purchase or a booking directly within the chat interface. By bringing in leadership that understands the intersection of user behavior and brand requirements, OpenAI is positioning ChatGPT as a full-funnel marketing tool—taking a user from the “awareness” stage to the “purchase” stage in a single conversation. Global Expansion: Canada, Australia, and New Zealand The current ad pilot has been largely focused on the United States, but that is about to change. OpenAI has revealed that it is actively exploring geographic expansion into Canada, Australia, and New Zealand. These markets are logical next steps for several reasons. First, they are predominantly English-speaking, which aligns with the current primary training and conversational strengths of the models. Second, these regions have high digital ad spend per capita and robust e-commerce infrastructures. By expanding into these “Tier 1” markets, OpenAI can test how its ad algorithms handle different cultural nuances and regulatory environments (such as varying privacy laws) before a broader global rollout across Europe and Asia. For international brands, this expansion is a signal to begin preparing creative assets and marketing strategies specifically for LLM (Large Language Model) environments. Advertising in a chat interface is fundamentally different from advertising on a static search results page, and brands in these upcoming regions will have the opportunity to be the first to solve that creative puzzle. User Experience and the “Quality Picture” One of the biggest risks for OpenAI is “ad fatigue” or the potential for ads to degrade the user experience. ChatGPT’s meteoric rise was fueled by its clean interface

The Transformation of the AI Interface: ChatGPT Ads Are Showing Up – A Lot For nearly two years, ChatGPT existed in the public consciousness as a clean, minimalist alternative to the cluttered, ad-heavy experience of traditional search engines. Users flocked to the platform not just for its generative capabilities, but for the relief of receiving a direct answer without having to navigate through sponsored results and pop-ups. However, that era of “pure” AI interaction is rapidly evolving. OpenAI has officially integrated advertising into its free tier, and the frequency of these placements is higher than many industry experts anticipated. What began as a quiet pilot program for free-tier users in the United States has blossomed into a full-scale monetization strategy. Recent testing and user reports indicate that advertisements are now a common fixture in the ChatGPT mobile experience, signaling a massive shift in how the world’s most popular AI platform intends to balance its astronomical operational costs with user accessibility. The Data Behind the Rollout: How Frequent are ChatGPT Ads? Recent investigations into the frequency of these advertisements provide a sobering look at the new user experience. In a controlled test involving 500 unique questions posed to the ChatGPT mobile app, researchers found that roughly 20% of new conversation threads triggered an advertisement. This means that for every five questions asked, users can expect to see at least one sponsored link appearing at the bottom of the AI’s response. These advertisements do not appear as traditional banner ads or intrusive pop-ups. Instead, they are presented as website link buttons. While they are visually distinct from the AI’s generated text, their placement is strategic, appearing immediately following the answer to ensure maximum visibility. This high frequency suggests that OpenAI is moving quickly to establish an ad-supported ecosystem that rivals the density found in social media feeds or traditional search engine result pages (SERPs). Understanding the Targeting Mechanism: Beyond the Keyword Traditional digital advertising relies heavily on keywords or browsing history. ChatGPT ads, however, leverage the unique nature of conversational AI. The targeting is based on three primary pillars: the specific topic of the current question, the user’s past chat history, and the information stored in ChatGPT’s “memory” feature. This contextual targeting makes the ads feel more integrated—and perhaps more persuasive—than traditional display ads. If a user asks for advice on training a puppy, the AI might serve an ad for a specific brand of premium dog food. If the conversation shifts toward business travel, the platform responds with links to hotel booking sites or corporate credit card offers. This level of granular targeting is a goldmine for advertisers, but it also raises new questions about how much personal data is being leveraged to fuel the ad engine. The Role of ChatGPT Memory in Advertising The “Memory” feature was originally designed to make the AI more helpful by remembering user preferences, such as a preferred programming language or dietary restrictions. Now, that same feature serves as a foundational component of OpenAI’s advertising platform. By utilizing memory, OpenAI can serve ads that aren’t just relevant to the immediate question, but to the user’s broader lifestyle and long-term interests. For example, if you mentioned a month ago that you were planning a wedding, the system might trigger ads for honeymoon destinations even when you are asking a seemingly unrelated question about vacation time. A Deep Dive into Ad Categories: From Travel to Tech The range of advertisers already participating in the ChatGPT ecosystem is surprisingly broad. Early data shows that certain industries are leaning into this new channel more aggressively than others. Travel remains one of the most prominent sectors; asking for help planning a trip to a specific location, such as Palm Springs, often triggers a Booking.com ad that automatically initializes a search for hotels in that specific geography. Other frequently spotted categories include: SaaS and Productivity: Tools for project management and workflow automation. Retail and Pet Care: Specifically high-intent items like specialized pet foods. Entertainment: Streaming services and tickets for live sporting events, such as basketball games. Financial Services: Corporate credit cards and accounting software targeting professional users. AI and Development: Coding tools and other AI-assisted software. This variety suggests that OpenAI is not just targeting casual consumers but is also positioning itself as a platform for B2B (business-to-business) marketing. The Rise of “Poaching”: A New Front in Brand Competition One of the most controversial dynamics emerging from the ChatGPT ad rollout is what marketing experts call “poaching” or brand conquesting. This occurs when a user mentions a specific brand by name, and the AI serves an ad for a direct competitor. For example, a query about DoorDash might trigger an ad for a different food delivery service, or a question about Netflix might lead to a promotional link for a rival streaming platform. This is a tactic long used in Google Search Ads, where brands bid on their competitors’ keywords. However, the conversational nature of AI makes poaching feel more direct. When a user asks an AI for help with a specific service, the appearance of a competitor’s link can disrupt the user’s intent and divert potential revenue. For brands, this means that even if they are the subject of a positive AI conversation, they are no longer safe from competitive interference within the same interface. OpenAI’s Stance: Balancing Revenue and Trust OpenAI is acutely aware of the potential backlash that comes with introducing ads into a platform that was once ad-free. To mitigate user concerns, the company has established several core principles for its advertising model. First and foremost, OpenAI maintains that ads do not influence the actual content of ChatGPT’s answers. The generative text remains unbiased (in theory), with the sponsored content restricted to the link buttons at the bottom. Additionally, the company has stated that full conversation transcripts are not shared directly with advertisers. Instead, the system uses the context of the conversation to trigger the ad without exposing the user’s full dialogue to third parties. Early

The Explosive Growth of OpenAI’s Advertising Ecosystem In the world of digital advertising, benchmarks are usually measured in years or decades. However, OpenAI has shattered traditional timelines by reaching a monumental milestone in record time. Just six weeks after launching its initial advertising pilot, ChatGPT has achieved over $100 million in annualized ad revenue. This rapid scaling comes at a time when the tech industry is closely watching OpenAI’s transition from a research-focused entity into a commercial powerhouse prepared for an eventual public offering. The significance of this $100 million figure cannot be overstated, especially when considering the controlled nature of the current rollout. Unlike established platforms like Google or Meta, which have mature, global ad infrastructures, OpenAI is still operating with what many would consider the “training wheels” of an advertising business. Yet, the early data suggests that ChatGPT is not just another platform for brands to park their budgets; it is becoming a high-performance engine for engagement and conversion. Breaking Down the Numbers: Scaling with Minimal Exposure Perhaps the most startling aspect of OpenAI’s $100 million revenue milestone is how little of the platform’s total capacity is actually being utilized. According to recent disclosures, this revenue was generated from less than 20% of eligible free-tier and “Go” tier users in the United States seeing ads on a daily basis. For advertisers, this signals a massive reservoir of untapped inventory that is set to come online in the coming months. Currently, approximately 85% of users on the Free and Go tiers are eligible to see advertisements. The fact that the current revenue is being pulled from a small fraction of that eligible base indicates that OpenAI is being extremely cautious with its user experience. They are prioritizing the “health” of the conversation over immediate profit, yet the profit is arriving regardless. This “fractional revenue” model suggests that as OpenAI ramps up the frequency and reach of its ad placements, the $100 million figure could easily quadruple or quintuple within the next calendar year. Currently, more than 600 advertisers are participating in the managed pilot program. These are largely enterprise-level brands working closely with OpenAI’s internal teams to test the efficacy of conversational ads. This curated approach has allowed OpenAI to monitor how users interact with sponsored content in a medium that is fundamentally different from a traditional search engine or social media feed. The April Shift: Self-Serve Access and Market Democratization For the broader marketing community, the most anticipated date on the calendar is now April. OpenAI has confirmed that it is on track to launch self-serve advertiser access during that month. This transition marks the end of the exclusive, managed pilot phase and the beginning of a competitive marketplace. The introduction of a self-serve platform is a pivotal moment in the history of any tech giant. It is the same move that allowed Google (then AdWords) and Facebook to scale from niche experimental tools into the dominant forces of global commerce. By opening the gates to small and medium-sized businesses (SMBs), OpenAI will likely see a surge in bid density and creative variety. Self-serve access means that local businesses, specialized e-commerce brands, and independent service providers can finally bid for space within the ChatGPT interface. This democratization of access usually leads to a “gold rush” period where early adopters can secure high-quality leads at a lower cost-per-acquisition (CPA) before the market becomes saturated and prices inevitably rise. Strategic Leadership: Why the Dave Dugan Hire Matters Scaling an ad business to billions of dollars requires more than just high traffic; it requires a sophisticated understanding of ad tech, auction dynamics, and brand safety. To lead this charge, OpenAI has made a high-profile hire by bringing on former Meta ad executive Dave Dugan to head up ad sales. Dugan’s background at Meta is particularly relevant. During his tenure there, he helped navigate the transition from desktop to mobile advertising and oversaw the development of complex targeting tools that became the industry standard. His move to OpenAI suggests that the company is looking to build a robust, enterprise-grade sales organization that can compete directly with the “Big Three” of digital ads: Google, Meta, and Amazon. Under Dugan’s leadership, we can expect OpenAI to move beyond simple text-based ads into more integrated, “agentic” commerce experiences. The focus will likely be on making ads feel like a natural extension of the AI’s assistance rather than an interruption of the user’s workflow. Geographic Expansion: Moving Beyond the US While the initial pilot has been heavily focused on the United States, OpenAI is already looking at its next phase of geographic growth. The company is currently exploring expansion into Canada, Australia, and New Zealand. These markets are often used as testing grounds for Western tech companies due to their high digital literacy and similar consumer behaviors to the US market. Expanding into these regions will provide OpenAI with a more diverse data set regarding how different cultures and demographics interact with AI-driven ads. It also allows global brands to begin coordinating cross-border campaigns within the ChatGPT ecosystem. For marketers in these regions, the April self-serve launch and subsequent geographic rollout represent a critical window to establish a brand presence before the platform becomes a standard part of every media buy. The Quality Picture: Balancing Revenue and User Trust One of the primary concerns surrounding AI advertising is the potential for “hallucinations” or irrelevant placements that could frustrate users. If a user is asking a complex coding question or seeking emotional support, a poorly timed or irrelevant ad could destroy the utility of the tool. OpenAI appears to be hyper-aware of this risk. Internal metrics show that fewer than 7% of ads are currently rated by users as “low relevance.” This is a remarkably low figure for a nascent ad platform. In comparison, traditional display advertising often suffers from much higher rates of negative sentiment or “banner blindness.” The company’s focus on relevance over volume is a strategic play to maintain user trust. By utilizing

The landscape of artificial intelligence is shifting. For the past two years, users have enjoyed a relatively “pure” experience with ChatGPT—an interface defined by its clean, minimalist design and the absence of traditional monetization hurdles. However, the honeymoon phase of ad-free AI utility is officially drawing to a close. Recent data and user reports indicate that OpenAI has aggressively integrated advertising into its free-tier ecosystem, and the frequency of these placements is much higher than many anticipated. The Arrival of the ChatGPT Ad Ecosystem For several months, OpenAI has been quietly piloting an advertising model for free-tier ChatGPT users in the United States. What began as a subtle experiment has evolved into a robust rollout that is now expanding internationally to markets including Canada, Australia, and New Zealand. This move signals a significant pivot for OpenAI, a company that once positioned itself as a research-first entity but is now grappling with the immense compute costs required to keep its models running for hundreds of millions of users. Recent testing and anecdotal evidence from the digital marketing community suggest that these ads are not just present; they are pervasive. For those utilizing the mobile app without a Plus subscription, the experience of interacting with a chatbot is starting to mirror the experience of using a modern search engine, albeit with a more conversational twist. Frequency and Placement: The 20 Percent Rule How often are users actually seeing these ads? In a comprehensive test involving 500 unique questions across the ChatGPT mobile application, a startling pattern emerged. Roughly one in five questions—or 20% of interactions—triggered an advertisement. These ads are typically found at the bottom of a conversation thread, appearing immediately after the AI provides its response. Unlike the intrusive pop-ups or banner ads of the early web, OpenAI has opted for a “website link button” format. These buttons are often labeled with phrases that suggest a helpful next step, such as “Find a hotel” or “Explore deals.” While the format is relatively clean, the frequency is what has caught many users off guard. In a standard conversation involving ten prompts, a user could realistically expect to see two distinct calls to action from sponsors. Contextual Targeting: How OpenAI Matches Ads to Prompts The most sophisticated aspect of this new ad system is how the targeting is handled. OpenAI isn’t just throwing random products at users; the ads are hyper-contextual. The system analyzes the current question, the overall topic of the conversation, and the user’s history to serve an ad that feels like a logical progression of the chat. For example, travel-related queries have proven to be the most lucrative and frequent triggers. A user asking for recommendations for a weekend getaway to Palm Springs might receive a perfectly curated response about hiking trails and mid-century architecture, only to find a Booking.com button at the bottom that automatically initiates a search for hotels in that specific city. This level of deep integration suggests that OpenAI is leveraging its understanding of user intent to provide high-value leads to its advertising partners. The Broad Spectrum of ChatGPT Advertisers The range of brands currently appearing in ChatGPT threads is surprisingly diverse. It isn’t just limited to massive travel conglomerates. The pilot program has seen ads for: Travel and Hospitality: Hotel bookings, cruise vacations, and airline tickets. SaaS and Productivity: AI coding tools, project management software, and corporate credit cards. Entertainment: Streaming services and professional sports tickets (specifically basketball). Consumer Goods: Dog food, wellness products, and retail items. This variety indicates that OpenAI is building a horizontal ad platform capable of serving almost any industry. If a user is discussing their dog’s diet, they might see a pet food brand. If they are debugging code, they might see a sponsored AI-driven developer tool. The model is built to capitalize on the specific “moment” of user need. The “Poaching” Dynamic: A New Battlefield for Brands One of the most controversial tactics observed in the current rollout is what marketing experts call “poaching.” This occurs when a user mentions a specific brand in their prompt, but the ad served is for a direct competitor. For instance, if a user asks about the best shows currently available on Netflix, the ad button at the bottom might lead to a subscription page for a rival streaming service like Hulu or Disney+. Similarly, a query about DoorDash might trigger an ad for a competing delivery platform. This is a classic tactic from the world of Search Engine Marketing (SEM), where brands bid on their competitors’ names to steal traffic. Its migration to AI is a clear sign that OpenAI is preparing for a future where brand visibility in “answer engines” is just as competitive as it is in traditional search results. For brands, this creates a defensive necessity. Not only will they want to be mentioned in AI outputs, but they may also feel compelled to pay for ad space just to prevent their competitors from appearing at the end of a conversation about their own services. The Privacy Question: What Does OpenAI Share? With any advertising rollout on a platform as personal as ChatGPT, privacy concerns are at the forefront of the conversation. OpenAI has been proactive in addressing these concerns, though their explanations leave some room for scrutiny. The company maintains several core pillars regarding data privacy and ads: No Answer Influence: OpenAI insists that ads do not influence the actual text generated by ChatGPT. The AI’s “opinion” or recommendations are ostensibly separate from the sponsored link at the bottom. Restricted Data Sharing: The company claims that the full content of a user’s conversation is not shared with advertisers. Instead, the targeting is handled internally. The Role of Memory: Ad targeting is based on the current question, past chats, and information stored in ChatGPT’s “Memory” feature. This means that if you told the AI three weeks ago that you were planning a wedding, it might use that context to serve you an ad today, even if your current

Search Engine Optimization (SEO) is frequently categorized as a long-term play, often siloed away from the immediate, data-driven world of Pay-Per-Click (PPC) advertising. However, the most sophisticated digital marketers recognize that these two channels do not operate in a vacuum. Instead, they form a symbiotic relationship where a high level of SEO maturity provides the essential infrastructure for paid media to reach its full potential. In a recent and highly successful marketing initiative for a B2B medical device company, this synergy was put to the ultimate test. By shifting the focus from short-term wins to building deep topical authority and medical trust, the brand achieved a staggering 133x Return on Ad Spend (ROAS). This case study explores the granular details of how SEO maturity served as the catalyst for unprecedented performance in a high-ticket, high-consideration market. The Challenge: Why Traditional Performance Playbooks Fail in Medical B2B Marketing a premium medical device, such as a specialized pelvic floor chair, is vastly different from selling consumer electronics or software-as-a-service (SaaS). In the medical sector, the stakes are exceptionally high, and the sales cycles are notoriously long. This is a classic “Your Money or Your Life” (YMYL) niche where Google’s standards for Experience, Expertise, Authoritativeness, and Trustworthiness (E-E-A-T) are at their most stringent. The target audience for these devices includes gynecologists, urologists, physiotherapists, and fitness center owners. These are highly educated professionals who do not make impulsive purchasing decisions based on a clever ad copy. They require clinical evidence, peer validation, and a sense of long-term reliability. At the start of this project, the brand faced several hurdles common to companies that treat SEO as an afterthought: Lack of Topical Authority: The website was not recognized as a leading voice in pelvic health. High Friction: Without a recognizable brand presence in organic search, PPC ads felt intrusive rather than helpful. Data Gaps: Incomplete tracking meant that marketing teams were “flying blind,” unable to see which touchpoints were actually driving sales. In this environment, simply increasing ad budgets or testing landing page colors was not enough. To scale, the brand needed to build a foundation of trust that would make every dollar spent on ads work significantly harder. Phase 1: The Initial State of Paid Media By the end of 2023, the brand had launched its first Google Ads campaigns focused on lead generation. While these early efforts did yield some sales, the infrastructure was fragile. At this stage, several critical issues were identified: Fragmented Tracking and Attribution Conversion tracking was rudimentary. Most conversions were being attributed to the “Direct” channel because the path from an initial search to a final sale was long and complex. Without clearly defined events in Google Tag Manager (GTM), the marketing team couldn’t see the nuances of user behavior. Furthermore, relying on GA4-imported conversions resulted in delayed signals, making it impossible for Google’s automated bidding algorithms to optimize in real-time. The “Cold” Outreach Problem Because the brand lacked organic visibility, every click on a paid ad was essentially a “cold” interaction. Users were being introduced to a high-ticket medical device for the first time through an ad. Without the reinforcement of organic search results or educational content, the conversion rate remained lower than desired, and the cost-per-acquisition (CPA) was high. However, these early campaigns served one vital purpose: they confirmed that search demand existed. The data showed that professionals were indeed searching for pelvic floor solutions. The problem wasn’t a lack of demand; it was a lack of brand authority. Phase 2: Treating SEO as Revenue Infrastructure In mid-2024, the strategy pivoted. SEO was no longer treated as a side project but as the core revenue infrastructure. The goal was to build a “trust layer” that would support all other marketing channels. This required a top-of-funnel educational strategy designed to capture users early in their research phase. Mapping the Informational Landscape Using Semrush, the team mapped the entire informational landscape surrounding pelvic health. This wasn’t just about targeting “buy” keywords. It was about answering the questions that doctors and patients were actually asking. The strategy focused on: Mechanism of Action: How does the technology actually work? Comparative Analysis: How does this chair compare to traditional physiotherapy or surgery? Clinical Evidence: Providing easy access to studies and medical whitepapers. Content That Educates Rather Than Sells The content strategy moved away from aggressive sales pitches. Instead, the brand invested in long-form, authoritative articles. These pieces featured structured FAQ sections and embedded videos featuring professional physiotherapists. By providing genuine value and clinical clarity, the brand began to be perceived as a partner in patient care rather than just another vendor. The Authority Lever: Partner-Driven Backlinks In the medical world, who you associate with is just as important as what you say. The most impactful SEO move during this phase was the development of a partner-driven backlink strategy. The brand already had a network of clinics and medical practices using their technology. The marketing team leveraged these existing relationships to build high-authority links that would be nearly impossible for a competitor to replicate. The Value Exchange The brand provided their partner clinics with high-quality, ready-to-use content. This included clinical study summaries, performance marketing visuals for the clinics’ own B2C lead generation, and educational blog posts. In return, the clinics linked back to the manufacturer’s website from their dedicated service pages. These were not generic directory links. They were contextual, highly relevant references from established medical domains. This strategy achieved two things: It passed significant “trust” and authority to the brand’s domain in the eyes of search engines. It placed the brand at the center of a specialized medical ecosystem, reinforcing its position as the industry standard. SEO Outcomes: Dominating the Search Results By the end of 2024, the results of this infrastructure-first approach were undeniable. The website began ranking #1 for critical generic terms such as “Beckenbodenstuhl” (German for pelvic floor chair). Beyond traditional rankings, the brand achieved dominance in AI Overviews and featured snippets. This organic dominance changed the psychology

The landscape of B2B digital marketing is currently undergoing a seismic shift. For years, the gold standard for lead generation was a robust keyword-driven strategy. Marketers would meticulously segment their Google Ads into brand and non-brand campaigns, bidding on high-intent terms and hoping for a steady stream of Marketing Qualified Leads (MQLs). However, if you are noticing that your performance is plateauing or that your Cost Per Acquisition (CPA) is climbing while lead quality drops, the problem likely isn’t the Google Ads platform itself—it is a strategy that has failed to evolve alongside the buyer’s journey. In the modern era, AI-forward campaigns like Performance Max (PMax) and Demand Gen are no longer just “experimental” options; they are the engines of future growth. But there is a catch that often scares off traditional B2B marketers: these campaigns require a level of patience and data-feeding that many organizations aren’t prepared for. If you are willing to move beyond the immediate gratification of the search bar, you will find a growth gold mine. Here is why AI-forward campaigns are essential and how to navigate the long road to success. The Evolution of the B2B Discovery Path The traditional marketing funnel assumes a linear path: a user has a problem, they search for a solution on Google, they find your ad, and they convert. In reality, the B2B buying process is far more chaotic. Modern buyers don’t start with a search engine when they are in the early stages of problem-solving. Instead, they live in ecosystems. They are researching pain points on Reddit, asking for peer recommendations in Slack communities, watching technical demos on YouTube, and increasingly “asking” AI tools like ChatGPT or Claude for vendor comparisons. By the time a prospect actually types your brand name—or even a category keyword—into Google Search, they have likely already formed an opinion about your product. If your strategy is entirely focused on capturing that final search, you are missing the 90% of the journey where the decision was actually made. You aren’t driving demand; you are merely trying to harvest it. AI-forward campaigns allow you to insert your brand into those earlier research phases across the entire Google ecosystem. Understanding AI-Forward Campaigns: PMax and Demand Gen Google has spent the last several years moving away from manual keyword management and toward multi-channel, multi-asset automation. Two primary campaign types lead this charge: Performance Max and Demand Gen. Performance Max (PMax) Performance Max is a goal-based campaign type that allows advertisers to access all of their Google Ads inventory from a single campaign. This includes Search, YouTube, Display, Gmail, and Maps. Instead of bidding on a specific keyword, you provide Google with “Audience Signals”—data points like your customer lists or specific interests—and the AI finds people who look like your best customers, regardless of which corner of the internet they are currently browsing. Demand Gen While PMax is focused on conversions across the entire funnel, Demand Gen is specifically designed to drive interest on Google’s most visual and immersive surfaces: YouTube (including Shorts), Discover, and Gmail. For B2B companies, this is where you can showcase customer testimonials, product walkthroughs, and thought leadership. It is the “top-of-funnel” engine that feeds the rest of your ecosystem. The beauty of these campaigns is their cost-effectiveness. In a traditional Search campaign, you might pay a premium to bid on a competitive non-brand keyword. In an AI-forward campaign, you might reach that same decision-maker while they are watching a relevant video on YouTube or scrolling through their Discover feed, often at a fraction of the cost of a Search click. The 4S + Ask Framework: Where Your Customers Live To succeed with AI-driven marketing, you must understand the “4S” framework of consumer behavior, which has been a staple of Google’s strategic advice. However, in the age of generative AI, we must add a fifth element: “Ask.” Search: Traditional intent-based queries on Google. Scroll: Passive discovery on social feeds, LinkedIn, and Google Discover. Stream: Consuming long-form or short-form video content on YouTube. Shop: Comparing prices, features, and reviews across platforms. Ask: Engaging with LLMs like Gemini or ChatGPT to synthesize information and get direct answers. If your B2B strategy only addresses “Search,” you are invisible during the “Scroll,” “Stream,” and “Ask” phases. AI-forward campaigns are designed to bridge these gaps. When a user “scrolls” through their feed, they see your display ad. When they “stream” a tutorial, they see your video ad. By the time they “search,” your brand is already the trusted authority in their mind. This holistic visibility is what builds the brand equity necessary to close complex B2B deals. Why Patience is the Ultimate B2B Competitive Advantage The biggest hurdle to adopting AI-forward campaigns in B2B is the “sales cycle hump.” In B2C, a user might see an ad for sneakers and buy them within ten minutes. In B2B—especially in sectors like SaaS, life sciences, or manufacturing—the time from first touch to closed-won can be six months, a year, or even longer. When you launch a Performance Max campaign, the initial data often looks discouraging. You might see a lot of impressions and clicks, but very few immediate conversions. At this stage, many marketers panic. They see the spend going up without an immediate return on ad spend (ROAS) and decide to pause the campaign, concluding that “AI doesn’t work for our niche.” This is a mistake. AI requires a learning period, and in B2B, that learning period is tethered to your sales cycle. Consider a case study of a life science company: their account managers almost killed a PMax campaign after three months because the platform data looked “soft.” However, they decided to wait. As the months rolled by and sales data began to flow back into the system, they realized that the PMax campaign was actually the primary driver of their highest-value contracts. The leads were discovering the brand via YouTube ads, researching for three months, and then finally converting via a branded search. Without the initial AI-driven

Introduction to the Google Ads API v23.2 Release In the rapidly evolving landscape of digital advertising, the ability to automate, scale, and gain granular insights into campaign performance is what separates market leaders from the rest. Google has recently announced the release of version 23.2 of the Google Ads API, marking another incremental but significant step in the platform’s journey toward more transparent and automated advertising solutions. This update is particularly relevant for developers, data scientists, and agency reporting teams who rely on the API to manage complex accounts and build proprietary tools. While some API updates focus on broad architectural changes, v23.2 targets specific “blind spots” that have persisted in the Google Ads ecosystem—most notably within Performance Max and App campaigns. By introducing new resources like VideoEnhancement and AppTopCombinationView, Google is providing the programmatic community with the data necessary to evaluate the efficacy of AI-driven creative elements. As the industry moves further into an AI-first era, these technical bridges allow advertisers to maintain a degree of human oversight over machine-generated outputs. Enhancing Transparency in Video Creative with VideoEnhancement One of the most discussed updates in the v23.2 release is the introduction of the VideoEnhancement resource. For years, one of the primary critiques of Performance Max (PMax) and other automated campaign types has been the “black box” nature of creative assets. Google often takes existing images, text, and video clips provided by an advertiser and uses generative AI or automated editing tools to create new video assets. While this helps fill inventory gaps, advertisers often struggle to report on whether a specific impression was served using their hand-crafted video or a Google-generated version. Understanding Advertiser-Provided vs. Google-Generated Content The VideoEnhancement resource now surfaces whether a video ad is Google-generated or advertiser-provided. This is a vital distinction for brand-conscious organizations. Many high-end brands have strict creative guidelines and want to ensure that their message is conveyed exactly as designed. With this new API functionality, developers can build reporting dashboards that explicitly flag AI-enhanced videos. By programmatically identifying these assets, agencies can now answer critical questions: Are Google-generated videos outperforming original assets? Do auto-enhanced videos maintain the brand’s aesthetic standards? This level of visibility allows for a more nuanced conversation between media buyers and creative teams, as they can now quantify the value of automated video generation within the Google ecosystem. Implications for Performance Max Reporting Performance Max has often been criticized for its lack of granular reporting compared to traditional Search or Display campaigns. The addition of VideoEnhancement data to the API is a direct response to the demand for more clarity. It allows for a more sophisticated analysis of the “asset group” performance, giving users the power to see exactly which components are being manipulated by Google’s algorithms to drive conversions. Optimizing Mobile Growth with AppTopCombinationView Mobile app marketing continues to be a massive vertical for Google, and v23.2 introduces a new tool for those managing App campaigns: the AppTopCombinationView resource. This new resource provides read-only insights into the top-performing asset combinations within App campaigns. App campaigns are notoriously automated, with Google’s machine learning deciding which combination of headlines, descriptions, images, and videos will resonate most with a specific user segment. Historically, getting a clear view of which specific “recipe” of assets was winning the most auctions was difficult to extract via the API. Leveraging Asset Combinations for Creative Strategy With AppTopCombinationView, developers can now programmatically retrieve the winning combinations. While the resource is read-only—meaning you cannot change the combinations directly through this specific view—the data it provides is invaluable for informing future creative production. If the data shows that a specific short-form video paired with a “Play Now” call-to-action is consistently the top performer, creative teams can lean into that style for their next set of assets. This update bridges the gap between raw performance data and actionable creative strategy. By pulling this data into third-party visualization tools, app marketers can provide stakeholders with a clear visual representation of what their most successful ads actually look like to the end user. New Control Settings for Demand Gen Campaigns Demand Gen campaigns, which replaced Discovery campaigns, are designed to capture user interest across YouTube (including Shorts), Discover, and Gmail. For travel and hospitality advertisers, Google has included a specific update in v23.2: the ability to disable the hotel feed via the HotelSettingInfo.disable_hotel_setting field. Granular Control for Travel Advertisers In previous iterations, managing how hotel feeds interacted with Demand Gen campaigns could be cumbersome. There are scenarios where a travel brand might want to run a broad brand awareness campaign using Demand Gen without necessarily pulling in the dynamic price-and-inventory feed from their hotel center. By providing a programmatic toggle to disable this setting, Google is giving developers more control over how specialized data feeds are utilized in cross-channel campaigns. This change reflects a broader trend in the Google Ads API: providing “opt-out” mechanisms for automated features that may not align with every advertiser’s specific strategy. It allows for a more tailored approach to campaign architecture, ensuring that the automation serves the advertiser’s goals rather than the other way around. Expanding Conversion Metrics: Indirect First In-App Installs Measurement remains the cornerstone of digital advertising, and v23.2 introduces a new metric that addresses the complexities of the modern app ecosystem. The API now supports a conversion metric for “indirect first in-app installs” across Campaign, Customer, and AdGroup resources. Navigating the Attribution Challenge In the world of mobile apps, attribution is rarely a straight line. Users may interact with multiple ads, visit a website, and eventually download an app through a path that isn’t a direct “click-to-install.” The indirect first in-app install metric helps capture these more complex conversion paths. For high-volume app advertisers, this metric is crucial for understanding the holistic impact of their ad spend. It allows for a better understanding of how Top-of-Funnel (TOFU) awareness campaigns contribute to eventual app downloads, even if those downloads aren’t immediately attributed to a direct ad click. By integrating this into the API,

Google has officially introduced a significant update to its Performance Max (PMax) campaigns, introducing a new feature called Asset Group Theming. This update is designed to help advertisers quickly adapt their creative assets for seasonal peaks, cultural holidays, and promotional events without the need for extensive manual redesigns or campaign overhauls. For digital marketers, the seasonal transition has traditionally been one of the most labor-intensive periods of the year. Swapping out summer imagery for fall aesthetics or preparing for the high-intensity Black Friday and Cyber Monday window often requires weeks of coordination between creative teams and account managers. With the introduction of Asset Group Theming, Google is leveraging its generative AI capabilities to bridge the gap between efficiency and creative relevance. Understanding the Shift in Performance Max Creative Management Performance Max has always relied heavily on the diversity and quality of its asset groups. Since PMax serves ads across YouTube, Display, Search, Discover, Gmail, and Maps, the system needs a wide variety of headlines, descriptions, images, and videos to find the best-performing combination for any given user. However, refreshing these assets for every minor or major holiday has been a persistent friction point. In the past, if an advertiser wanted to pivot from a general “Always On” campaign to a Valentine’s Day or Back to School theme, they had two main options: replace the existing assets (which risks losing historical performance data) or build entirely new asset groups from scratch. The new Asset Group Theming feature offers a third, more streamlined path by allowing advertisers to apply seasonal “wrappers” to their existing high-performing creative structures. How Asset Group Theming Works The core mechanic of this update is built around the concept of “cloning and skinning.” Instead of starting with a blank slate, the tool allows advertisers to take a successful asset group and apply a specific theme. Google’s AI then analyzes the existing images and text to generate themed variations. When a theme is applied, the AI typically focuses on the following adjustments: 1. Image Background Modification The tool uses existing product or lifestyle images as a base and modifies the background or surroundings to fit the chosen theme. For example, a product shot used in a summer campaign could be automatically updated with a snowy background or festive lighting for a winter holiday theme. This allows for visual consistency while signaling relevance to the current season. 2. Textual Suggestions In addition to visual updates, the system suggests headlines and descriptions aligned with the theme. If a “Sale” theme is applied, the AI might suggest phrases like “Limited Time Offer” or “Seasonal Savings,” integrating them into the existing ad copy structure. It is important to note that these are suggestions and typically only a handful of lines are updated at once, ensuring the core messaging of the brand remains intact. 3. Safe Testing Environment One of the most critical aspects of this feature is that it leaves the original asset group untouched. By cloning the group before applying the theme, advertisers can run the new seasonal version alongside the original or pause the original while the holiday season is active. This protects the “learning” and historical data of the primary assets, making it easier to revert once the season ends. Comprehensive List of Available Themes Google has launched this feature with a wide array of themes that cover the most significant retail and cultural events globally. These are categorized into three primary buckets: Promotional, Seasons, and Cultural Moments. Promotional Themes These themes are designed for specific sales cycles rather than a calendar date. Sale: Focuses on urgency and value-driven messaging. Studio/Editorial: Provides a more polished, high-fashion, or minimalist look for brand-heavy campaigns. Seasonal Themes These are broad themes used to align the “vibe” of the ad with the current time of year. Winter: Cool tones, snow, and cozy indoor settings. Spring: Floral elements, bright lighting, and themes of renewal. Summer: High saturation, outdoor activities, and sun-drenched aesthetics. Fall: Warm earth tones, autumn leaves, and preparation for the colder months. Cultural and Holiday Moments This is the most granular category, covering specific holidays that drive massive spikes in consumer spending. Black Friday/Cyber Monday: High-impact, high-urgency themes for the peak shopping weekend. Christmas and Hanukkah: Traditional festive decor and gift-giving imagery. Halloween: Spooky or autumn-themed creative elements. Valentine’s Day: Romantic and gift-focused aesthetics. Easter: Pastel colors and spring-related holiday symbols. Mother’s Day and Father’s Day: Themes focused on appreciation and gifting for parents. New Year and Lunar New Year: Celebration, fireworks, and themes of new beginnings. Back to School: Education-focused imagery and preparation for the academic year. Strategic Benefits for Advertisers The rollout of seasonal creative theming isn’t just a convenience; it’s a strategic tool that addresses several common pain points in the Google Ads ecosystem. Reducing Creative Friction In many organizations, the bottleneck for launching a new campaign is the creative department. Designers are often spread thin, and requesting 20 different aspect ratios for a single holiday can take days or weeks. Asset Group Theming allows account managers to generate a “v1” of seasonal creative in minutes. This can serve as a placeholder while custom assets are being built or as a permanent solution for smaller brands with limited design resources. Preventing Creative Fatigue Creative fatigue occurs when an audience sees the same ads so often that they stop clicking or, worse, develop a negative association with the brand. By quickly rotating themes—moving from a general “Summer” look to a “Back to School” look—advertisers can keep their presence fresh in the eyes of the consumer without changing the underlying product offering. Agility in Market Response Consumer trends move fast. If a particular cultural moment gains traction, brands that can pivot their creative quickly often see a higher Return on Ad Spend (ROAS). The ability to apply a theme with a few clicks allows brands to be more reactive to the calendar than they ever were with manual asset management. Important Limitations and Human Oversight While the AI-powered nature of this