OpenAI's Ad Gambit: A Stopgap on the Road to Agentic Commerce?

OpenAI's announcement that it would begin testing ads on ChatGPT marks a pivotal inflection point for the AI giant. According to Business Insider, Evercore ISI analyst Mark Mahaney projects that advertising could become a $25 billion annual business for OpenAI by 2030. That sounds bullish until you look beneath the surface.

The reality is stark: OpenAI is hemorrhaging money at a pace rarely seen in tech history. The company's burn rate has reached approximately $9 billion annually, and it expects cumulative cash burn of $115 billion through 2029.  

For context, competitor Anthropic expects to break even by 2028, with its burn rate projected to drop to roughly one-third of revenue in 2026 and just 9% by 2027. OpenAI, by contrast, expects its burn rate to remain at 57% in 2026 and 2027. The company expects to burn through roughly 14 times as much cash as Anthropic before turning a profit in 2030.

This isn't a company leisurely exploring new revenue streams. This is a company that needs cash, and needs it now. The ads announcement is less a strategic pivot than an acknowledgment of financial gravity.

The Google Irony

The irony here is worth noting: OpenAI is not the first company dragged into advertising against its original philosophy. The original reluctant advertiser? Google itself.

In their 1998 Stanford research paper, "The Anatomy of a Large-Scale Hypertextual Web Search Engine," Larry Page and Sergey Brin explicitly warned that "advertising funded search engines will be inherently biased towards the advertisers and away from the needs of the consumers." They argued that superior search would actually reduce the need for ads. Yet Google became the most successful advertising company in history, generating nearly $300 billion in ad revenue in 2025 from Search and YouTube alone.

Now OpenAI finds itself in the same position: a company built on the promise of intelligence-first interactions, contemplating whether to litter that experience with sponsored content.

Clayton Christensen's Framework

This brings us to what Clayton Christensen termed "The Innovator's Dilemma." In his 1997 work, Christensen demonstrated how successful companies can do everything "right" and still lose their market leadership. The core insight: established firms optimize for their existing customers and revenue streams, making them vulnerable to disruptive technologies that initially seem inferior or irrelevant.

Google is living this dilemma in real time. The company could have beaten OpenAI to the generative AI punch. It had the talent, the compute, and the research (Transformer architecture originated at Google, after all). But Google was reticent to test generative technology aggressively because doing so would cannibalize its search advertising revenue. Why encourage users to get answers directly from an AI when you profit from them clicking through multiple search results?

This hesitation created the opening OpenAI exploited. Although Google is playing catch-up, shareholders cannot fault the company from making hay when the sun shined - cashing in on ad-driven search was the only rational play in pre-GenAI world. Now is a different story. Google has launched  subscription services like Google AI Pro at $26.99/month and Google AI Ultra at $339.99/month (CAD). The fact that Google is experimenting with subscription models at all suggests the company recognizes its advertising cash cow may have a finite lifespan.

The Streaming Precedent

OpenAI and Google aren't alone in their reluctant embrace of advertising. The streaming industry provides a cautionary tale.

Netflix, which famously built its brand on ad-free viewing, launched its ad-supported tier in late 2022. By 2025, the company generated over $1.5 billion in advertising revenue and projects that figure to double to approximately $3 billion in 2026. Amazon Prime Video followed suit in January 2024, instantly becoming the largest ad-supported subscription streaming service in the world. By late 2025, Prime Video reached 315 million monthly ad-supported viewers globally.

The pattern is clear: companies that promised premium, uninterrupted experiences eventually succumb to the siren song of advertising revenue. The question isn't whether ads compromise the user experience. The question is whether the alternative (running out of cash) is worse.

Beyond Ads—The Agentic Commerce Model

Advertising may be OpenAI's stopgap solution, but it is unlikely to be its endgame.

The Walmart Signal

In October 2025, Walmart announced a partnership with OpenAI to create what both companies call "agentic commerce." The collaboration allows customers to shop directly through ChatGPT using Instant Checkout. As Walmart CEO Doug McMillon put it: "For many years now, eCommerce shopping experiences have consisted of a search bar and a long list of item responses. That is about to change."

This is the real signal. OpenAI isn't just thinking about displaying ads alongside chat responses. It's positioning itself as an intermediary between consumers and retailers, a position that carries far more revenue potential than advertising.

The "Costco Model" for AI

Consider what happens as agentic AI matures. You might tell ChatGPT: "Order my usual groceries from Walmart for pickup on Saturday, but check if there are any good deals on chicken this week. And remember, I'm still doing keto."

In this scenario, OpenAI becomes something like a Costco for the AI age: a membership-based service where you pay for access to automated, intelligent commerce. The value proposition isn't just the AI itself but the integrations, the reliability, the human-in-the-loop quality assurance during the early phases, and eventually, the pure automation.

This model offers multiple revenue streams:

• Consumer memberships: Users pay a monthly fee for access to premium agentic services
• Merchant fees: Retailers like Walmart pay for preferred integration status
• Transaction fees: A small percentage of each completed purchase

However, the Costco analogy has limits. Costco's model derives 73% of its gross profit from membership fees, which work because the company leverages massive purchasing power to negotiate wholesale pricing from suppliers. OpenAI would lack this kind of supplier leverage; its value would come from convenience and AI intelligence rather than from negotiating better prices. A more accurate framing might be that OpenAI would function as a digital concierge service with membership economics, not a wholesale negotiator.

The Third Wave of Commerce

We've seen commerce evolve from physical stores to e-commerce. Agentic AI represents a third wave where computation doesn't just facilitate your purchase, it makes the purchase for you. OpenAI and Anthropic could bypass both Amazon's retail dominance and Google's search dominance simultaneously by becoming the trusted intermediary between consumers and merchants.

The real money isn't in showing you ads for products. It's in being the system that handles your entire purchasing relationship with the world.

Conclusion

OpenAI's move into advertising is understandable given its current burn rate, but it should be viewed as a bridge, not a destination. The company needs cash to survive long enough to build something more durable. That something is likely agentic commerce: a membership-based model where AI companies act as trusted intermediaries, guaranteeing accuracy, handling customer service, and eventually automating the entire consumer-merchant relationship.

Google warned against ad-funded search in 1998 and became an advertising colossus. Now OpenAI, built on the promise of direct intelligence, may follow the same path, at least temporarily.

It's also worth noting that we're in the early phases of this transition, and major retailers are hedging their bets. In January 2026, Walmart announced a similar partnership with Google, allowing customers to shop directly through the Gemini app. This suggests that even as agentic commerce takes shape, the ultimate winners remain unclear, and the largest retailers are positioning themselves to work with whichever AI platform prevails.

The question for OpenAI isn't whether ads will generate revenue. The question is whether OpenAI can execute fast enough on the agentic commerce vision before burning through its capital or compromising the user experience that made ChatGPT dominant in the first place.

Author: Malik D. CPA, CA, CISA. The opinions expressed here do not necessarily represent UWCISA, UW,  or anyone else. This post was written with the assistance of an AI language model. 

Five Top Tech Takeaways: A Deep Dive into Disruption, From AI Hardware to GenAI-Powered Services

Author's Note: In this post, we explore five top tech stories, focusing on disruption and how it relates to the humane AI pin, Rabbit R1, and GenAI-powered innovations.

The humane AI pin and Rabbit R1 aim to disrupt the smartphone market, but they face challenges in delivering the convenience and functionality users demand. Disruption, as Harvard professor Clay Christensen explains in his books "The Innovator's Dilemma" and "The Innovator's Solution," is not just about introducing new technology but rather about offering a compelling product at a price point that attracts customers who have been ignored by the market. Overserved customers who find the extra features in current product lines unnecessary will gravitate towards cheaper, convenient, and "good enough" alternatives. 

HP's success with inkjet printers exemplifies the concept of disruption. Inkjet printers offered a cost-effective alternative to laser printers while outperforming dot-matrix printers in terms of quality. This innovation met the needs of users who did not require the advanced features of laser printers but sought better quality than dot-matrix printers could provide. As a result, inkjet printers successfully disrupted the market by catering to the demands of overserved customers.

GenAI-powered agents, note-taking apps, and intelligent search provide helpful features that make people's lives easier but are affordable. Take, for example, Dr. Lall's use of an AI-enabled note taker (see below). It is an excellent example of how this technology can effectively "amplify" one's effort. As I discussed in my Medium post, GenAI has the potential to be leveraged as a junior assistant, capable of drafting emails, conducting research, and performing other content-oriented tasks typically assigned to a remote virtual assistant. Consequently, these innovations can potentially disrupt the market by focusing on delivering the functionality users need at a more accessible price point.

MKBHD on Humane AI Pin: An Overhyped, Underdelivering Gadget

Marques Brownlee's review of the Humane AI Pin, a wearable AI gadget, exposed its primary flaw: it's not inherently flawed technology but rather a flawed concept. Despite the tech's shortcomings, such as slow processing and inaccuracy, the real issue lies in attempting to replace the smartphone with a device offering less functionality. Users aren't looking for a chest-worn gadget that can't match their phones, making the AI Pin an impractical solution that ultimately fails to deliver on its promise. The broader lesson here is that not all innovative ideas are worthwhile, and Humane's assumption that people need an alternative to smartphones was misguided.

Key Takeaways:

• Marques Brownlee criticized the Humane AI Pin for being an underwhelming alternative to smartphones with less utility and greater inconvenience.
• Despite its design team's good intentions, the AI Pin's limited features and inability to integrate with common tools like Google Calendar reveal its conceptual flaws.
• The AI Pin’s attempt to replace smartphones fell short, demonstrating that new tech ideas must address real user needs to be successful.

The Rabbit R1: Promising Concept, Disappointing Execution
The Rabbit R1, a new AI gadget priced at $199, fails to deliver on its promise of offering an advanced and practical AI assistant. The device is plagued by incorrect identifications, faulty integrations, and limited capabilities, particularly with apps like Uber and Spotify, which struggle to execute basic functions. Though its Large Action Model (LAM) is meant to simplify tasks across different apps, the current implementation is unreliable and often frustrating. While its whimsical design and quality microphone provide some appeal, the Rabbit R1 largely underwhelms, suggesting that AI gadgets still lag far behind smartphones in utility.

Key Takeaways:

• The Rabbit R1's AI capabilities, such as food and object identification, are unreliable and often incorrect.
• Integrations with apps like Uber, DoorDash, and Spotify are poorly executed and offer limited practical use.
• Despite its attractive design, the Rabbit R1 struggles to compete with the functionality and reliability of smartphones.

(Source: The Verge)

Amplification in Action: How AI is Saving Time for Family Physicians

Dr. Rosemary Lall, a family physician in Scarborough, Ontario, discovered a groundbreaking solution to the overwhelming administrative burden that nearly drove her to quit her practice. By implementing an artificial intelligence note-taking application called AI Scribe, Dr. Lall significantly reduced the time spent on mandatory patient record-keeping. The AI Scribe, developed by OntarioMD, automatically generates detailed SOAP notes during patient visits, which has drastically cut down on the after-visit paperwork and allowed Dr. Lall to focus more on patient care and less on administrative tasks.

Key Takeaways:

• Ontario doctor adopted AI Scribe to address the extensive administrative duties that were impacting her work-life balance.
• AI Scribe assists in creating SOAP notes, thereby reducing paperwork and saving time for physicians.
• The Ontario government is conducting a pilot program to integrate AI Scribe into more practices, indicating a move towards broader adoption of AI in healthcare management.

(Source: Global News)

OpenAI's Sam Altman Foresees an AI Assistant Revolution

In an interview with MIT Technology Review, OpenAI's CEO Sam Altman discussed his vision for AI tools that will significantly integrate into our daily lives, much like smartphones. He envisions AI as a “super-competent colleague” that can manage various tasks seamlessly, adapting to users' needs and learning from interactions without feeling intrusive. While he doubts that new hardware will be necessary for this paradigm shift, Altman suggests consumers might still appreciate a specialized device. He remains optimistic about overcoming the challenges of sourcing training data for future AI models and anticipates that multiple versions of AGI will excel in different areas. He hinted at ongoing development of future models but declined to disclose specifics about GPT-5's release date.

Key Takeaways:

• Sam Altman sees the "killer app" for AI as a highly capable virtual assistant that can tackle tasks independently.
• Despite the challenges in sourcing training data, Altman remains hopeful about finding new methods for advancing AI capabilities.
• Altman anticipates several versions of AGI that will vary in their abilities but did not provide a timeline for GPT-5.

(Source: MIT Technology Review)

Apple Plans to Bring AI to iOS: Intelligent Search and Beyond

Apple is preparing to introduce significant enhancements to its Safari web browser, including integrating an AI-powered tool called Intelligent Search. Set to launch with Safari 18, alongside iOS 18 and macOS 15 later in 2024, these updates aim to enhance user experience through advanced content blocking, a Web Eraser feature for removing specific webpage elements, and AI-driven content summarization capabilities. These improvements are part of Apple's broader strategy to implement more secure and efficient AI technologies in response to the growing influence of generative AI tools in the tech industry.

Key Takeaways:

• Apple's Safari 18 will feature Intelligent Search, utilizing on-device AI for advanced browsing and text summarization.
• The Web Eraser tool in Safari will enable users to selectively erase webpage content, enhancing privacy and user control.
• Apple continues to align its software capabilities with AI advancements, positioning Safari as a more competitive and secure browser option.

(Source: AppleInsider)

Author: Malik Datardina, CPA, CA, CISA. Malik works at Auvenir as a GRC Strategist who is working to transform the engagement experience for accounting firms and their clients. The opinions expressed here do not necessarily represent UWCISA, UW, Auvenir (or its affiliates), CPA Canada or anyone else. This post was written with the assistance of an AI language model. The model provided suggestions and completions to help me write, but the final content and opinions are my own.