The future is female: Why AI must centre women’s health

By Lynda Wilkes-Green, Founder of Ahlya

As artificial intelligence becomes more deeply embedded in healthcare, we’re standing at a pivotal crossroads where we can either continue to build technology that reflects historic gender bias or we can consciously design a future that recognises and responds to the realities of women’s health.

For too long, women have been underrepresented in the data that informs medical decisions.

From clinical trials that favoured male participants to doctors who dismiss women’s symptoms as “normal,” the result is a persistent gender data gap that skews diagnostics, treatments, and outcomes, and too often leaves women without answers.

This is not just a statistical oversight, it’s a systemic issue with very real consequences.

The Invisible Woman in Health Data

Historically, women’s health has been treated as a niche concern.

Until 1993, women were routinely excluded from clinical trials in the US – a policy only formally changed by the National Institutes of Health Revitalization Act, which mandated the inclusion of women and minorities in all NIH-funded research.

Even now, many medical studies do not analyse results by sex, and women living with complex or misunderstood conditions like endometriosis, PMDD, and PCOS continue to face misdiagnoses, delayed care, and medical gaslighting.

AI promises to personalise healthcare like never before, but it is only as inclusive as the data it’s trained on. If we’re not actively collecting and centring female data, we risk building a new wave of tools that repeat old mistakes, but this time, at scale.

The Rise of Generative AI in Wellness

In the past year, we’ve seen a surge in generative AI being applied to wellness, from mental health chatbots to cycle tracking apps.

These tools offer speed, scalability, and hyper-personalised insight; but without gender-aware design and development, they also risk simplifying or misrepresenting female physiology.

Take menstrual health, for example, many apps still reduce the cycle to a calendar or ovulation predictor, ignoring the nuanced interplay of hormones that affect everything from mood and energy to sleep and appetite.

For women with irregular or anovulatory cycles, which are common with conditions like PCOS, these tools often fall short or deliver inaccurate recommendations.

We need a new standard.

We need to develop AI that is aware of hormonal complexity and designed to support wellbeing across the entire cycle, not just fertility, and move beyond prediction models towards dynamic, lived-data insights.

Designing with Hormones in Mind

Centering women’s health in AI is not just about fairness; it’s about effectiveness.

Hormones influence nearly every system in the body, from brain function to metabolism, yet they are rarely accounted for in mainstream health tech.

By integrating cycle data with daily mood, symptom, and behaviour tracking, we can begin to build models that respond to how women actually feel, rather than just what the textbook says.

This opens the door to more empathetic, adaptable tools that meet users where they are, especially during the luteal phase, which is a time when many women experience mental health dips, fatigue, and emotional sensitivity that often go unsupported.

For users with irregular cycles or conditions like PCOS, personalisation is even more critical. A

I can offer meaningful support by learning patterns over time, offering insights not tied to fixed timelines but to symptom clusters, emotional rhythms, and lifestyle triggers.

Ethics, Trust, and Representation

Innovation without integrity is not enough.

If we are asking women to share intimate data, from cycle logs to mental health patterns, we must uphold the highest standards of transparency, ethics, and security.

Too many femtech products still lack clear data governance policies. In some cases, user data has been shared with third parties without informed consent.

These breaches of trust erode the very foundation that responsible AI must be built on.

Inclusive AI also requires inclusive teams.

We need more women at the table when products are being built. Representation matters, not just in marketing, but in design, development, and leadership too.

A Call to Collaborate

AI offers us the tools to radically transform women’s health, but only if we use them consciously.

Closing the gender data gap is not a solo project; it requires cross-disciplinary collaboration between founders, researchers, clinicians, and policymakers.

It means listening to women and building systems that evolve with the complexity of female bodies, rather than simplifying them for convenience.

As we enter a new era of health innovation, let’s not replicate old patterns.

Let’s create something better, something that sees women, hears women, and builds with women in mind.