Contraceptive App Algorithm Evolves Again: Natural Cycles’ New AI Cleared by FDA

The red or green indicator displayed each day by a contraceptive app may look like a simple interface choice, but it involves a high-stakes judgment: when pregnancy may be possible, and when abstinence or additional protection such as condoms is still required. The U.S. Food and Drug Administration (FDA) recently cleared Natural Cycles’ updated fertility algorithm, moving the software-based contraceptive tool further toward personalization.

The FDA database shows that Natural Cycles Nordic AB’s submission K253694 received 510(k) clearance on July 17, 2026, and was classified as a Class II “fertility diagnostic and contraceptive software application.” This decision means the FDA determined that the updated device is substantially equivalent to a legally marketed predicate product; it does not mean the regulator has comprehensively endorsed all claims about the AI’s performance.

The app integrates daily body temperature, menstrual cycle dates, and optional user-entered ovulation or pregnancy test results to determine whether a given day is a “Green Day” or a “Red Day” requiring additional protection. The company says the new NC° Fertility Algorithm combines physiological modeling with machine learning and was trained on tens of millions of real-world fertility data points. Its goal is to adapt more quickly to changes in individual cycles and improve the handling of irregular cycles and lower-quality physiological signals.

Natural Cycles says the validation submitted to the FDA showed that the new algorithm maintained existing contraceptive effectiveness while enabling many users to receive more Green Days. The company continues to cite overall figures of 98% effectiveness with correct use and 93% with typical use. However, the announcement did not specify the number of participants or cycles in the validation of this update, the population composition, the actual increase in Green Days, or a separate performance range for people with irregular cycles. It is therefore currently difficult to assess the magnitude of the improvement and the boundaries of its applicability based on publicly available information.

The practical significance of this update lies in the need for software-based contraceptive tools to balance safety and usability. If an algorithm is overly conservative, users will receive more Red Days and convenience will decline. If it labels Green Days too aggressively, it may misclassify fertile periods as low risk. Ovulation timing can vary particularly with illness, stress, the postpartum state, or discontinuation of hormonal contraception. Whether machine learning can maintain reliable performance amid these changes is more important than simply adding personalized labels.

Natural Cycles also said it has filed patent applications covering some of the new algorithm’s technology. This is the product’s sixth FDA clearance since it first received De Novo authorization in 2018; previous updates have also covered wearable-device integration and prescription and over-the-counter availability. This regulatory development confirms that the updated software may be marketed in the United States, but its true clinical value will still require more complete validation data and real-world postmarket performance among users with different cycle patterns to clarify.