Bloomberg Surveillance · Thursday, September 10, 2026
Heath Terry of Citi notes that while open-source AI models have gained traction, recent frontier model releases are proving to be not only performant but also economically efficient on a cost-per-task basis. He explains that despite potentially higher per-token costs, these new models consume fewer tokens, making them more cost-effective for enterprises. Terry believes this efficiency will lessen the concern around open-source competition going forward.
“Open source was always going to be a big part of AI the same way it was a big part of software. And we're seeing a lot of open source growth, not just from Chinese models, but from all of the Western models as well.”
“And what we're seeing now is that these newest models are actually a lot more efficient. And so even though their cost per token is higher, They burn fewer tokens to answer the question or complete the query or the task that's been assigned to it. So the cost per task is actually lower.”
“And I think that ultimately is what's going to matter at the enterprise. So I don't want to say that we're completely past open source being a concern, but I do think it's going to be significantly less of a concern going forward.”