Artificial Intelligence in In Vitro Fertilisation: Clinical Evidence, Market Trajectory, Regulation and Bioethics, with Foresight to 2031
Year: 2026
Type: article
Abstract: Research monograph (working paper, not peer reviewed), 58 pages, 52 references. Background. Around one in six people experience infertility, more than three million IVF cycles are reported annually and an estimated 13 to 17 million children have been born through IVF. Artificial intelligence (AI) is now applied along the entire IVF pathway while its clinical benefit remains contested. Aim. To provide an integrated, critical assessment of AI in IVF as of late 2026 across clinical evidence, market development, regulation and ethics, and to derive scenario-based predictions and recommendations. Methods. Structured narrative review, documentary analysis of EU, Swiss, UK and US regulation, secondary analysis of market data, normative ethical analysis and scenario analysis. Results. AI embryo assessment is roughly ten times faster than manual assessment, but the largest randomised trial (n = 1,066) did not demonstrate noninferiority on clinical pregnancy (46.5% vs 48.2%), and time-lapse imaging did not improve live birth in a trial of more than 1,500 patients. The AI embryo selection market is estimated at USD 320 million in 2025, growing about 18.5% annually. Under the EU AI Act most clinical IVF AI is high-risk. Polygenic embryo screening is considered not ready for clinical practice by the ASRM. Conclusions. The study proposes the cohort constraint as an explanatory principle, a four-question governance framework for clinics, ten falsifiable predictions to 2031 and a research agenda. AI should inform and accelerate; accountable humans should decide. Use of AI: an AI language model assisted with literature synthesis, structuring and language editing; the author verified all sources and takes full responsibility.
Author: Berky, Nathaniel Jonathan
Language: English
Subfield: Health Informatics
Field: Medicine
Domain: Health Sciences
Open Access status: green
OpenAlex ID: W7236727265
