#62: OpenAI’s Monetization Roadmap
Commerce, media, and advertising. And hopefully a lot more
We are beginning to see OpenAI turn on the jets of its monetization strategy.
Last week, we chatted about the OpenAI x Shopify partnership, where users can purchase directly from Shopify-based brands via Instant Checkout, the new ChatGPT commerce interface. And now Walmart has joined the Instant Checkout party, announcing a partnership with OpenAI, through which ChatGPT users will be able to purchase from the e-commerce goliath within a prompt.
We’re soon headed for a world where you can buy everything in ChatGPT. It reminds me of the concept of a “Super App”, particularly of China-based WeChat, which is a singular platform for social, commerce, and banking. But here, you can have your essay edited and buy paper towels on the same platform.
Now, as neat and efficient as this scenario sounds, I do have doubts about how much new shopping demand will be created (given consumer spending constraints) and believe that most of Instant Checkout sales will be a share shift from existing e-commerce platforms to OpenAI. The new front-end UX will be technologically impressive; however, that doesn’t mean consumers suddenly have more money to buy things.
Following the “Super App” analogy, it seems that OpenAI is looking to build the “do everything” interface. Or that is, the “do everything that OpenAI has a clear path to monetize.”
People go to ChatGPT for guidance on work tasks, school assignments, travel itineraries, and even for therapy (maybe they shouldn’t). These types of tasks can (and have been) monetized via a freemium model. A free version of ChatGPT is available, but the bread-and-butter benefits of the app exist behind a subscription paywall. I imagine ChatGPT subscription prices will increase over time for consumers and businesses. But there’s a realistic cap on how high they can raise their prices given competition from other AI companies. And that cap won’t quench investors’ thirst who invested at a $500B valuation.
Commerce
So, the next logical thing is, ChatGPT introduces commerce. They have plenty of data on what people are searching for and the intent behind their search. It’s no surprise that Walmart partnered with OpenAI, as SimilarWeb data suggests that 20% of Walmart.com referral traffic comes from ChatGPT. Kind of crazy, huh? If you were skeptical about the impact AI is having on purchase discovery, your doubts have been laid to rest.
Commerce itself is a trillion-dollar industry. Given the grand ambitions of OpenAI and the sheer amount of capital they have raised, OpenAI will need to disrupt the world’s biggest industries to back up their claims. Thus, it makes sense they zeroed in on allowing people to buy from within ChatGPT (via Instant Checkout).
Advertising
After commerce, I predict that advertising will be up next (and very likely is in the works right now). Because of the growing range of needs that ChatGPT can serve, you’ll inevitably spend more time within the platform, which means your attention is the next asset OpenAI will monetize. There’s only 24 hours in the day (unfortunately, AI can’t scale time), so every incremental minute that a company can capture your time from something else matters. Oh, and guess what? Advertising is a trillion-dollar industry too.
I imagine OpenAI will announce a few more commerce partnerships with e-commerce heavyweights (think Target, eBay, etc.) before they feel confident with their infrastructure. Then they’ll focus the news outlets on their next endeavor: advertising. My guess is that, by April 2026, you’ll see a “sponsored” suggestion next to the product you are looking to buy or the flight you plan to book. Also, OpenAI is able to recognize the intent behind a prompt. I wouldn’t be surprised if most transaction-related prompts have either Instant Checkout or an advertisement (or both) in the near future.
Media
How could I be remiss for not mentioning that OpenAI has disrupted the media industry, a trillion-dollar industry in its own right? The Sora text-to-video app is the wildest application of AI that I have seen to date. And by wildest, I mean it allows you to create complete nonsense. Write a prompt for a video idea and Sora will generate a video of surreal quality (check out a few examples here). The video generation capabilities of Sora will make you question reality and will continue to blur the lines between authentic, human-generated content and AI.
Although the content created in Sora is considered by many critics to be “slop,” the technological advancements in text-to-video have to send a chill down the spine of anyone working in Hollywood. It seems plausible that we are not too far away from writing a prompt that generates an entire movie or sitcom for you. Soon you’ll be able to watch Ocean’s Eleven with yourself as the lead instead of George Clooney. There will be many intellectual property (IP) issues as this wave of content personalization emerges, but I bet that innovation will work around some of them. Hollywood studios will be able to produce top-tier content at a fraction of the price today thanks to advancements in AI.
As we’ve spoken about before, distribution is powerful and arguably as important as a great product. Here’s a bold prediction: OpenAI will buy Disney by the end of 2026. Disney has the IP and OpenAI have the way to personalize old content and make new content at pennies on the dollar of current studio budgets. By acquiring into the distribution engine, OpenAI circumvents a lot of IP lawsuits and turns on the jets on media monetization through the Disney umbrella of properties.
Software
Another big area I haven’t yet touched on is how AI is disrupting software production, one of the largest industries today. OpenAI is certainly a major player in disruption, but Anthropic has branded itself as the AI model company for code generation. Although not an engineer, I rely on Anthropic’s Claude (and sometimes Google’s Gemini) for guidance when writing SQL queries in my day-to-day.
The grandeur vision of OpenAI spans far beyond an e-commerce marketplace, an advertising platform, or a media outlet. However, these are low-hanging-fruit industries for OpenAI because AI has already proven out that people look for guidance on what to buy from AI and are willing to spend a lot of time within the interface. Not only will we be likely to see improvements in customer experience thanks to AI in the way of hyper-personalized product recommendations, advertisements, and television, but also these avenues kickstart the revenue machine that is table stakes to meet investor requirements. Capitalism is quite a beast, huh?
Commerce, advertising, and media. The trillion-dollar trifecta. But what about other industries? What’s next on OpenAI’s roadmap?
It remains to be seen. But a blurb from last Friday’s edition of The Rundown gives me hope that AI can be useful for things besides monetizing our attention and taking more money out of our wallets.
OpenAI hired a black hole scientist as the first academic researcher for its OpenAI for Science division. Although the announcement in The Rundown was terse, I find it energizing that the AI behmoths are prioritizing scientific research throughout their monetization push. This type of work can be monetized in the long run, particularly through disrupting the healthcare industry, but isn’t going to lead to a short-term bump in revenue.
I believe the AI that will make our lives better won’t be workflow automations or relevant ads. It’ll be breakthroughs in medicine and climate, allowing us to live healthier and longer in a sustainable world.
I’m bullish about AI for many reasons, including those applications that aren’t focused primarily on monetization.