In a September 8 company essay, Sarah Friar argues that stronger models and more efficient computing can make additional work economically viable. She connects OpenAI’s consumer products, enterprise adoption and infrastructure investment to its plans for growth.
The essay describes OpenAI’s experience with agents and its efforts to reduce the cost of serving models. Those are the company’s claims about its own operations. They do not establish the savings a creative studio, agency or brand would achieve, and the essay does not announce a new customer price cut.
For a team assessing AI spending, the relevant comparison is the cost of a finished, accepted deliverable. Include the time spent correcting outputs and reviewing them, alongside model charges. A vendor’s lower infrastructure cost is useful context, but a purchasing decision needs evidence from your own workflow.