Smart Labs, Real Questions: A Guide to AI in Fertility Treatment

In 2019, fewer than half of IVF clinics in the U.S. used any form of artificial intelligence in their embryology labs. By 2025, the FDA had cleared the first AI-powered embryo assessment tool, a startup had announced the birth of babies conceived through fully automated IVF, and headlines were declaring that robots were “picking” embryos. If you’re mid-treatment or about to start, the question isn’t whether this technology exists. It’s whether any of it actually matters to your cycle.

The short answer: some of it is promising, some of it is overhyped, and almost none of it works the way the headlines suggest. Below, we break down what AI and robotics are actually doing in IVF labs right now, what the research says about whether it helps, and what you might want to ask your doctor.

First, the basics. When people say “AI is selecting embryos,” they’re almost always talking about one specific task: embryo grading. After eggs are fertilized in a lab, embryologists evaluate the resulting embryos to determine which ones look most likely to lead to a pregnancy. Traditionally, this has been done by a trained embryologist looking through a microscope and assigning a grade based on visual characteristics: how symmetrically the cells are dividing, whether the outer layer (the trophectoderm) looks healthy, how the inner cell mass is developing.

AI tools like Fairtility’s CHLOE and Life Whisperer do something similar, but using algorithms trained on thousands of embryo images. They analyze visual patterns that may be too subtle for the human eye and produce a score or ranking. Some systems work with time-lapse incubators (cameras that photograph the embryo every few minutes as it develops), building a kind of developmental movie that the AI can analyze frame by frame.

The key thing to understand: AI isn’t making the decision about which embryo to transfer. It’s giving your embryologist another data point. Think of it like a spell-checker for embryo grading; it flags things and offers suggestions, but a human is still reading the document.

This is where it gets complicated, and where the honest answer matters more than the optimistic one.

A systematic review published in Human Reproduction Open found that AI models achieved a median accuracy of about 81.5% when predicting embryo viability, compared to roughly 51% for embryologists using standard visual grading alone. That sounds like a dramatic improvement. But accuracy in a study isn’t the same thing as better pregnancy rates in a clinic.

When researchers ran an actual randomized controlled trial, the results were more sobering. A 2024 study published in Nature Medicine tested an AI scoring system called iDAScore against traditional morphology-based selection in over 1,000 patients. The pregnancy rate in the AI group was 46.5%; in the standard group, 48.2%. The AI system was not shown to be even equivalent to the traditional approach, let alone better.

A separate meta-analysis from 2024 pooled data across multiple studies and found that AI’s ability to predict implantation had a sensitivity of 0.69 and specificity of 0.62, with an overall predictive score (AUC) of 0.70. In plain language: helpful, but far from a crystal ball. The researchers flagged major validation gaps, meaning most AI tools haven’t been tested widely enough across different clinics, patient ages, and ethnic backgrounds to know how consistently they perform.

The robotics side of the story is newer and smaller, but it’s real. A company called Conceivable Life Sciences developed a system called AURA that can perform many of the manual steps in IVF, including sperm injection (ICSI) and embryo transfer. In a small study, five babies were born from twelve automated embryo transfers, a 41.7% live birth rate. One procedure was even performed remotely, with the operator controlling the robot from 2,300 miles away.

As MIT Technology Review reported, these are proof-of-concept results, not standard practice. The sample sizes are tiny. But the vision behind the technology matters: IVF labs are facing a serious staffing problem. According to industry estimates, the world will need 43,000 more embryologists and 4,000 new IVF labs by 2035 just to meet current demand. Automation could help stretch limited expertise further, potentially reducing costs and expanding access to clinics that can’t recruit enough specialists.

That’s the real promise of robotics in IVF: not replacing the people who care for your embryos, but making high-quality lab work possible in more places.

Not everyone in reproductive medicine is enthusiastic about the pace of AI adoption, and for good reason.

The American Society for Reproductive Medicine (ASRM) released a statement in late 2025 concluding that polygenic embryo screening (a newer approach that looks at hundreds or thousands of genetic markers simultaneously to estimate an embryo’s risk for complex conditions like heart disease or psychiatric disorders) is “not ready for clinical practice and should not be offered as a reproductive service at this time.” Unlike standard genetic testing, which checks for specific single-gene conditions like cystic fibrosis, polygenic screening tries to predict risk for diseases caused by many genes interacting together. The concern: polygenic risk scores don’t perform equally across different ancestries, the gene-environment interactions are poorly understood, and the ethical implications of screening embryos for conditions like depression haven’t been adequately addressed.

A 2025 paper in Human Reproduction mapped out the ethical landscape more broadly, identifying concerns about algorithmic bias (AI trained mostly on data from certain populations may not work as well for others), the “black box” problem (clinicians and patients can’t see how the AI reaches its conclusions), and what the authors called “machine paternalism”; the risk that an opaque algorithm could quietly shape reproductive decisions without meaningful informed consent.

Researchers at Monash University echoed these concerns, noting that the conversation about AI in embryo selection has moved faster than the evidence supporting it. “The technology is being marketed to patients before we fully understand its limitations,” the team cautioned.

If your clinic uses AI-assisted embryo grading, that’s not a red flag. The tools that have reached clinical use are designed to supplement, not replace, your embryologist’s judgment. They may help your team make a more informed ranking of embryos, especially in cycles where multiple embryos reach the blastocyst stage and the differences between them are subtle.

But it’s also not a reason to choose one clinic over another, or to assume that AI-assisted selection will meaningfully change your odds. The largest randomized trial to date didn’t show a benefit. The technology is evolving quickly, and future versions may perform differently, but right now, the human expertise of your embryology team matters more than which software they’re running.

AI and robotics are genuinely changing the IVF lab. Some of those changes will likely improve outcomes over time; others are still catching up to their own hype. The headlines that say AI is “picking your embryo” are skipping past the part where a trained embryologist reviews the data, considers your specific medical history, and makes a judgment call informed by years of experience.

The technology is a tool. A promising one, with real limitations, being used by real people who are trying to help you build your family. That’s worth knowing, and it’s worth asking about. But it’s not worth losing sleep over.

Disclaimer: This article is for informational purposes only and is not intended as medical advice, diagnosis, or treatment. Every fertility journey is different, and the information here should not replace a conversation with your doctor or a qualified reproductive health professional. Always consult your healthcare provider before making decisions about your fertility, starting or stopping any treatment, or trying new supplements, devices, or wellness tools. If you’re concerned about your fertility, a reproductive endocrinologist can provide guidance tailored to your specific situation.