Odd Lots · Thursday, July 9, 2026
A recent paper from Bridgewater discussed their use of proprietary data to fine-tune an open-source AI model for identifying newsworthy financial information. The study found that combining open-source models with proprietary data yielded superior results compared to frontier AI models.
“Actually totally. There was just one more thing. I don't know if you caught it last week. Were recording this July seventh, last week while you're on vacation. It was a very interesting paper out for Bridgewater talking about their use of proprietary data to fine tune an open source version of Quinn and for a specific purpose of being able to identify what is newsworthy financial information and not. They found that the combination of open source models plus proprietary data got them better results than the most frontier US models.”