• There are no suggestions because the search field is empty.
Woman sitting at a table at home, looking at her smartphone

A focus merely on speed alone won’t land you in the winner’s circle.

Healthcare costs keep climbing, and employers and their people are paying the price. Employers expect health benefit costs to rise 8.2% per employee in 2027, the biggest jump since 2003. Women’s health sits at the center of that challenge: maternity costs are now a top-three cost driver for one in five large employers. And AI, as it’s currently deployed, isn’t necessarily helping: in fact, most large employers surveyed believe it’s driving up near-term costs.

As the spotlight shines on Las Vegas this week for HLTH 2026, this year’s theme, “The Age of Intelligent Healthcare Is Here,” makes me wonder...are we really thinking intelligently about AI?

News coverage and press releases have highlighted that much of AI has been narrowly aimed at speed and so-called efficiency gains: faster calls, faster claims, faster answers. Faster is better, the thinking goes. But better how? And for whom? Those of us who build and buy healthcare technology should be widening our lens to ask a different question: how can we use AI so that people get healthier, have a positive experience, and everyone spends less getting there?

Faster isn’t always better

“When faced with a difficult question, we often answer an easier one instead, usually without noticing the substitution." Daniel Kahneman, Thinking Fast and Slow.

Having my entire career based in Silicon Valley and the tech start-up world, I know how deeply the “move fast” instinct runs. So it’s no surprise speed became the default target for healthcare AI as well: cost deferment and short-term gains are tangible and easy to count. But that leads to the friction between investment, returns, and outcomes. Too often the investments are not lining up with what is best for the patient. It’s been reported that AI has cut nursing jobs from record review. It’s not that far-fetched of an idea that this could lead to “optimizing staff” by closing clinics. If we don’t make the right choices.

Today, the prevailing approach uses data to predict the “next action,” then automates or expedites it. The problem is that this approach treats every member call or interaction as a moment to be optimized, not an opportunity to deepen understanding, build trust, and change the trajectory of someone’s care for the better. Isn’t that why we’re in healthcare? To improve care for the better? Consider a member who has been trying to conceive for eight months. They call their fertility benefits provider with a simple question: how many cycles does the benefit cover? In one version, an agent with no record of earlier conversations confirms coverage, shares a list of clinics, and closes the ticket. The answer and the handling of the call is efficient, by any standard measure. In another version, the member’s dedicated care advocate answers the question, then asks how the last eight months have really been, and learns about irregular cycles, a relentless job, rising stress, and sleepless nights, and a diagnosis of endometriosis. The member leaves with a personalized plan and someone who knows their story.

This difference is not simply in the answer. It is in knowing what to ask next. An accurate answer can address the immediate question while leaving other needs undiscovered—not because the agent gave the wrong information, but because the person on the phone lacked the context or cues to explore further. AI can help bridge that gap by drawing on available information to prompt thoughtful follow-up questions, helping advocates better understand members’ circumstances, and guide them toward relevant care. Sometimes, the right simple question can open a much more meaningful conversation.

So, in this scenario, while the second call took a few minutes longer, it was also far more valuable for the member and for whoever pays for the care. Women’s health journey—from preconception through pregnancy and postpartum, and into midlife—stretch over months or years, through setbacks, emotional decisions and long waits. No single call can or should tell the whole story. Its value lies in what it adds to the relationship, not how efficiently it’s handled. This is where healthcare has something to learn from the world of hospitality. In Setting the Table, restaurateur Danny Meyer famously distinguished service, the technical delivery of a product, from hospitality, the way that service makes someone feel. By that standard, the measure for any member services team becomes not “how fast did we resolve the call?” but “did the member feel supported and understood, and are they closer to the best outcome?”

That’s why Progyny has never put its Progyny Care Advocates (PCAs)—the guides who support every step of a member’s care—on a clock. The job is not to shorten conversations but to make every call smarter and more valuable, while being efficient and respectful of our member’s time.

Trust is the foundation. AI should strengthen it.

Where AI can add enormous value is by making every conversation smarter, not faster. By carrying context forward, so a call starts not with “Who are you?” but with “How has [your concern] been since we last talked?” And by enabling personalization; not boxing every member into the same answer.

And AI should bring rich data to every conversation. At Progyny, that we have a decade of insight from guiding millions of people through their journeys—significantly more than anyone else. That’s what I think of as intelligence-informed care.

And that’s why at Progyny, we’ve built a whole new technology system that focuses on surfacing protocols AND personalization to the PCA, so that they have all the information that they need at their fingertips to make sure the conversation is fluid, complete, and helpful.

Every area of healthcare deserves AI that is working for the member. In the women’s health space in particular, we are proud to be using our data to building on the most talented, most expert clinicians in the women’s health space, and with the goal of delivering the best feedback loop available. To not seek to replace human connection, but rather arming them with better tools and systems to support care delivery; then keeping the loop going to continue to support more training. Not only are we not replacing them—but we are directly investing in their skills, experience and models of care. Investing in humans, to women and their health and family building around the globe.

AI should be in the background surfacing context and choices, but it’s the human who should make the call using their judgement. We know that 3 members can ask the same question and need 3 different things. A protocol treats them alike and makes sure nothing is missed—but that is table stakes for quality care. A PCA equipped with insight and empowered to use their judgment knows best and gives equal weight to the member experience.

The long game is the smarter investment

On the surface, all of this might sound expensive: longer calls, dedicated advocates, time spent on relationships. But it only looks that way if your measure is optimizing transactions. Women's health doesn’t work that way, and neither do its costs.

The best proof is that nearly half of Progyny members reach their family building goals without needing fertility treatment. A PCA is still assisting their journey, not trying to fix them into a box—this is goals and personalization over corporate protocol. For those who do need treatment, determining the right path early matters just as crucially. It’s by making this a priority that Progyny’s model delivers a 23% higher live birth rate, a 22% higher single embryo transfer rate and a multiples rate 61% lower than the national average. The wrong path, in contrast, means lost months, unnecessary cycles and riskier and more costly pregnancies—with multiples carrying 4.8 times higher NICU and delivery costs. Slowing the conversation down to listen to our members is often the fastest way to the best outcome, and the surest way to bend the cost curve.

Taking the measure of intelligence

The age of intelligent healthcare may well be here. But let's not gauge intelligence by how fast a system predicts the next action or how many calls it closes. In healthcare, the most intelligent AI is the kind that knows a patient best. It carries their story forward, pulls years of collective insight into a critical moment, and gives the people caring for them the time and information to do what only humans can.

That’s also how, heading into another “unprecedented” year, we’ll bend the cost curve. Lasting savings come from getting people on the right path early, with someone they trust beside them.

As we gather at HLTH, here’s my invitation to everyone building and buying healthcare AI: let’s raise the bar for what we call “intelligent,” and measure it by whether people end up healthier and better cared for.

Looking for more? Here are some related resources.

smiling couple relaxing and holding up their baby
How JM Family Enterprises connect associates to transformative fertility & midlife care
woman deep in discussion with HR at work
Get started now to bring a women’s health benefit to your company | Progyny
happy businesswomen working on laptops in the office
How Texas Capital responded to employee demand for comprehensive women's health benefits
woman presenting data on a screen at work
Mind the gap: new data reveals what’s missing in your women’s health benefits
What are the most cost-effective family building benefits for self-insured employers? | Progyny