OpenAI has begun letting some large customers pay only when its AI tools deliver results, according to a report from The Information. The shift signals a broader rethinking of software pricing as AI systems take on work that once required human employees.
The previously unreported arrangement lets clients tie payment to outcomes rather than flat subscription fees. The Information cited a source familiar with the matter in its Sunday report.
Outcome-based pricing spreads across software
Salesforce is adopting a similar approach with its Agentforce AI. Businesses can negotiate custom contracts that charge based on whether the AI grows revenue by helping salespeople close more deals or cuts costs by automating customer service interactions.
"Customers want to buy and want to price in different ways. This is something I've learned really aggressively recently," CEO Marc Benioff said on an investor call last week.
The move comes as startups like OpenAI launch outcome-based pricing models that pressure established software companies to adapt. The approach has complications, though. Stripe has issued guidance warning that successful outcomes "could stem from product tweaks, marketing campaigns or seasonality" rather than the software itself. Without explicit attribution rules, Stripe said, "customers could argue over whether the outcome belongs to you."
What this means for customer support teams
The per-seat pricing model worked for years because it aligned incentives across customers, vendors, and investors. AI agents complicate that picture. A support platform powered by AI may resolve a growing share of tickets autonomously, making per-agent charges less intuitive when much of the work is automated.
For customer support professionals, this pricing shift matters because it reflects how vendors now measure their products: by outcomes like resolved tickets and reduced costs, not by how many seats you buy. That could reshape contracts, vendor conversations, and how support teams justify their technology budgets.
Why this matters for customer support
If your company pays for AI support tools based on outcomes, expect sharper scrutiny of what the software actually delivers. Vendors may push for contracts tied to automated resolution rates or cost savings, which means support leaders will need clear data on what the AI handles versus what human agents still resolve. The ability to measure that split accurately could determine whether an outcome-based contract saves money or creates disputes.
Your membership also unlocks:
