OpenAI just dropped a bombshell earlier this year that they're looking to take a possible 2% revenue share of Shopify products recommended in Deep Research. This got me thinking. Could OpenAI be testing the waters for a bigger play in the B2B space? Could this be the groundwork for OpenAI to take a cut of a business's deal that was recommended through Deep Research? Something that could fundamentally alter how B2B transactions happen forever.
We're talking about an AI middleman taking a cut of business deals—even enterprise-level deals. All from a single prompt.
The Shopify Test Run
Let's be clear about what's happening. OpenAI has been open about starting to collect revenue share when people buy products through ChatGPT's Shopify integration via Deep Research. It's brilliant, actually. You ask ChatGPT for product recommendations, it shows you Shopify results, you buy, OpenAI gets paid. I walked through this user experience and it raised several questions of its own! (That will be tomorrow's hypothetical rabbit hole of the day)
But here's the thing—B2C is child's play compared to B2B. Consumer purchases are impulse decisions with immediate conversion tracking. B2B deals? They could be six-figure contracts that close months after the initial conversation, involve multiple stakeholders, and have zero traditional tracking mechanisms.
Is this what OpenAI wants? Imagine the revenue stream!
The B2B Revenue Share Rabbit Hole
Picture this: A startup founder asks ChatGPT, "What's the best ERP system for a 50-person SaaS company?" and they have a ChatGPT integration with one of these CRMs: Salesforce, HubSpot, or Zoho. The founder is given results that include SAP, NetSuite, and SageX3. The founder clicks on the SAP link and completes a contact form. Six months later, that founder signs a $100,000 annual contract with SAP.
What would OpenAI's revenue share model be on the business side? Let's use the Shopify revenue share for argument's sake. With a 2% B2B revenue share model, OpenAI could claim $2,000 from one conversion. Now multiply that across millions of business queries daily. The numbers get absurd quickly.
My initial thought is this will be a technical and attribution nightmare! However, imagine the business nightmare that has been created over $2,000.
The Technical Nightmare
Traditional affiliate tracking is useless here. No UTM parameters to track the old-school way. Server-to-server tracking is useless too! How would you take into consideration procurement processes? But OpenAI has something traditional affiliates don't: Unchecked CRM integrations.
ChatGPT already syncs with Salesforce, HubSpot, and Zoho. What if those integrations weren't just for productivity—what if they were tracking mechanisms? OpenAI could monitor when companies mentioned in AI conversations appear in CRM pipelines, when deals progress, when contracts close.
The AI doesn't need cookies when it's sitting inside your sales process.
The Profit Margin Apocalypse
Here's where this gets truly disruptive. B2B companies operate on margins, revenue forecasts, and budgets. Enterprise software might have 80% gross margins, but after sales, marketing, and operational costs, net margins often hover around 10-20%.
Now imagine paying OpenAI 2% of every deal they influenced. Companies would face an impossible choice: absorb the cost and destroy profitability, or raise prices and lose competitiveness. Either way, OpenAI wins while traditional businesses get squeezed.
It's the ultimate tax on doing business in an AI-powered world.
The Attribution Wars Begin
The real battle will be over attribution. When does OpenAI deserve credit for a deal? If someone asks about CRM options in January and signs a contract in June, does OpenAI get paid? What if they researched five other sources in between?
Smart businesses will try to game the system. They'll train employees to never mention vendor names in ChatGPT. They'll use alternative AI tools for research. They'll build internal knowledge bases to avoid AI recommendations entirely.
But here's the kicker—OpenAI doesn't need perfect attribution. They just need enough volume to make billions while everyone else argues about fairness. How could all of this be impacted by another outside source, like a company such as Cloudflare's paid AI indexing service?
The Endgame: AI as the Ultimate Gatekeeper
This isn't really about revenue sharing. It's about control.
Once OpenAI embeds itself in B2B decision-making, they become the invisible hand guiding every purchase. They decide which vendors get recommended, which solutions appear first, which companies get buried in search results.
Traditional sales and marketing become obsolete. Instead of building relationships with customers, businesses will need to optimize for AI algorithms. Instead of brand recognition, companies will need AI recommendation scores.
OpenAI won't just take a cut of existing deals—they'll control which deals happen at all. They have every keystroke of data and now combine that with the knowledge of what companies they make the most referral revenue from. Businesses might have to make the decision to opt for less automation and efficiency for their bottom line.
The Questions Nobody's Asking
While everyone debates AI safety and alignment, OpenAI is quietly positioning itself to tax the entire online economy. The questions we should be asking are:
The Uncomfortable Truth
OpenAI's B2B revenue share model is plausible, but is it inevitable? The technology exists, the integrations are live, and the financial incentives are massive.
The only question is whether businesses will recognize the threat before it's too late.
Because once the AI tax becomes standard, there's no going back. OpenAI will have successfully inserted itself into a vast majority of online transactions, taking a cut while providing nothing but algorithmic recommendations.
It's brilliant, ruthless, and exactly what you'd expect from a company that just convinced the world to pay for automated responses.
If the AI middleman era has begun, your profit margins will never be the same.