Scientists at Columbia University Fertility Center in the United States have used artificial intelligence (AI) to find viable sperm in a man who had been declared infertile for 18 years. The breakthrough has now helped the couple achieve pregnancy through IVF.
The man had been diagnosed with azoospermia, a condition where sperm is absent in semen. For nearly two decades, he and his wife had tried unsuccessfully to conceive.
Doctors had spent two days examining his semen samples but found no sperm. However, when the new AI system was used, it identified 44 live sperm within one hour.
Dr. Zev Williams, who led the research, said the discovery could change the future of fertility treatments. “Labs searched for two days and found nothing. We ran the same sample and found 44 sperm within an hour,” he said.
The technology, known as STAR (Sperm Track and Recovery), uses a special chip and high-speed camera to scan semen and identify live sperm using AI-trained software. These can then be used for IVF through a process called intracytoplasmic sperm injection (ICSI), where a single sperm is injected directly into an egg.
In March 2025, the couple had their first successful embryo transfer using sperm detected by the AI system. The woman is now pregnant.
Experts say the innovation reduces the need for painful surgeries to extract sperm from the testicles and could be life-changing for many couples.
However, some fertility experts have urged caution, warning that more studies are needed before the technology can be used widely.
