This is an episode of the Glossy Beauty Podcast, which features candid conversations about how today’s trends are shaping the future of the beauty and wellness industries. More from the series →
Subscribe: Apple Podcasts • Spotify
Oura Health, the Finnish wearables company that has sold more than 5 million health tracker rings, is betting on women’s health with the launch of its first-ever proprietary large language model designed specifically for women.
“We know historically that women have been underrepresented when it comes to a lot of [medical and pharmaceutical] research,” Tanvi Jayaraman, MD, clinical lead of health AI at Oura, told Glossy. “We want to change that narrative when it comes to women’s health.”
LLMs are the brains behind AI chatbots, including Oura’s in-app Advisor chat where users can ask general wellness questions, specifics about their personal health data or in-depth medical questions.
“Women have been searching for answers [about our health and bodies on the internet] for just as long as the research has been done,” she said. “The answers that [women are] looking for are really disparate and scattered. They’re on a niche Reddit forum, or they’re kind of word-of-mouth, so a lot of [what we learn online is] hypothesis-driven, data-gathering one-offs.”
Starting last year, Dr. Jayaraman’s team of board-certified clinicians began “training” Oura’s new LLM with only the best data and studies available. This is juxtaposed against many other LLMs, which are trained on the internet at large, which can result in hearsay and causality connections being learned as fact, Dr. Jayaraman said.
“[When we’re able to] pick and choose the right training data, the right sources, the right guidelines for women’s health, then you can start to push away some of that noise [from the internet],” she said. “Of course, we have a long way to go when it comes to the actual research, but you have to start somewhere.”
Dr. Jayaraman represents a new type of physician who bridges medicine, artificial intelligence and product strategy.
After medical school at Stanford, she worked on AI strategy projects at Bain & Company, working for global diagnostics and pharmaceutical companies, then on Apple’s clinical team, where she worked on next-gen digital health tools. She joined Oura last year.
Dr. Jayaraman joined the Glossy Beauty Podcast to discuss Oura’s new women-focused LLM, the future of AI-powered wellness chatbots and more.
On advocating for women’s health in tech
“A huge focus of mine during medical school was actually in women’s health. I led the Women’s Health Clinic at our free clinic over at Stanford for a year or so. I think what we see is that women’s health has long been underserved, and there’s kind of this one-size-fits-all data model that we see, and so we really wanted to change that with our first custom LLM. We designed it and vetted all of the training data with the clinical team of PhDs on staff and our currently practicing OB-GYN. We really wanted to make sure that everything this LLM was trained on was actually used in the real world by physicians today, so it’s data and knowledge that’s uniquely vetted by our in-house team. A big part of building this LLM was making sure that the data was actually usable by clinicians down the road. So we really want to make sure that women are getting the guidance and the support they need that is truly trustworthy. … We really feel like we’re pushing the boundaries of what’s possible with AI and with wearable health data coverage, and we hope that this is the blueprint for future potential use cases for other health conditions. I see the power of AI. I get really excited about it as being a way to connect the dots between what’s the most reliable, validated science with what’s actually happening in your body.”
On developing the personality of the new LLM-powered chatbot
“Yes, the knowledge base is extremely important, but we have also trained the AI responses and fine-tuned with real clinicians in the loop in the process. So it’s not enough for the AI to just be pulling from a certain knowledge base, but we also really need clinicians to evaluate every output and say, ‘OK, would I actually say it this way in the room?’ What is the structure and format? What’s the tone? Where does the empathy come in? What words do we use, being really intentional about every aspect? The evaluation of the AI has also been a critical differentiator, and it’s really important for us at Oura to make sure that we are keeping that clinician at the center of that process. So my role was really testing our AI, looking through thousands of AI outputs and giving feedback on what responses seemed aligned with what I would say in my clinic. … We were really building the AI and fine-tuning it based on this real-world experience of clinicians today. And that’s another layer that really helped us build a product that we can stand by.”
