If you've ever toggled between ChatGPT, Claude, and whatever other AI tool caught your eye this week, you're apparently keeping OpenAI's leadership up at night.
The Verge obtained a four-page internal memo sent by OpenAI's chief revenue officer, Denise Dresser, to employees over the weekend. The message is candid about where the company sees its biggest challenges - and it's not just about building smarter models. It's about keeping people around once they've found them.
The 'moat' problem
The memo reportedly hammers on one concept repeatedly: building a moat. In business terms, that means creating enough value, habit, or lock-in that customers don't feel the pull to leave. And in the AI space right now, that's genuinely hard to do. Switching from ChatGPT to Anthropic's Claude, for example, takes about thirty seconds and zero dollars. There's no contract to break, no data to migrate, no real friction at all.
That low switching cost is a structural headache for any company trying to build a dominant AI product - and Dresser's memo signals that OpenAI knows it. The strategic focus, according to the memo, involves not just attracting users but binding them more tightly to the platform through enterprise relationships and deeper product integration.
The enterprise push
Growing the business side of things appears to be central to the plan. Enterprise clients - companies paying for API access, custom deployments, or ChatGPT Team and Enterprise subscriptions - are far stickier than individual users. When a business builds its internal tools around one AI provider, migrating becomes a real project, not a casual afternoon experiment.
This shift in emphasis toward corporate clients makes sense given how competitive the consumer AI space has become. With Google, Anthropic, Meta, and a growing list of others all offering capable models, winning on raw performance alone is increasingly difficult.
What this means for regular users
For everyday ChatGPT users, this memo is a useful reminder that the AI tools you rely on are businesses with competitive pressures, not neutral utilities. The features being built, the pricing strategies being tested, the integrations being prioritized - they're all shaped by the goal of making you less likely to open a competitor's tab.
That's not necessarily sinister. Competition is generally good for users, and OpenAI working harder to earn loyalty could mean better products. But it's worth knowing what game is being played when a company says it's focused on your experience.
