In a case that feels like the plot of a science fiction novel, an eight-year-old rescue dog in Sydney named Rosie has become the first in the world to receive a personalized mRNA cancer vaccine designed by her owner using artificial intelligence. The experimental treatment, developed after conventional therapies failed, has successfully shrunk a tennis ball-sized tumor on Rosie’s leg by approximately 75%, stunning the scientific community.
The journey began in 2024 when Rosie, a Staffordshire Bull Terrier-Shar Pei cross, was diagnosed with aggressive mast cell cancer and given only months to live. Her owner, Paul Conyngham—a data engineer with no background in biology—refused to accept the prognosis. Leveraging 17 years of experience in machine learning, Conyngham turned to AI chatbots to map out a research strategy. Under the guidance of these tools, he spent $3,000 to have the DNA of both Rosie’s healthy tissue and her tumor sequenced at the University of New South Wales (UNSW).
Using Google’s AlphaFold to model protein structures and analyzing gigabytes of genetic data, Conyngham identified the specific mutations driving the cancer. When a pharmaceutical company denied a request for an existing immunotherapy drug, he collaborated with the UNSW RNA Institute to manufacture a bespoke mRNA vaccine based on his data-driven formula. After a grueling three-month process to secure ethical approval—which involved Conyngham drafting a 100-page document—the vaccine was administered at the University of Queensland’s School of Veterinary Science in December 2025.
The results were almost immediate. Within weeks of the first injection, researchers observed a dramatic reduction in tumor size. Beyond the physical shrinkage, Rosie’s overall quality of life improved significantly; her owner reported that after months of lethargy, she was recently seen jumping fences to chase rabbits at a local park.
While the “citizen science” aspect of the project has drawn praise from tech leaders like OpenAI’s Greg Brockman and DeepMind’s Demis Hassabis, medical experts urge a balanced perspective. Biomedical engineers note that while a single successful case is a remarkable “proof of concept,” it does not yet constitute a proven cure. However, the researchers involved believe Rosie’s case proves that personalized medicine can be developed in a time-sensitive manner, potentially offering a blueprint for future human oncology.
Conyngham is currently working on a second vaccine to target a remaining tumor that did not respond to the initial round, hoping to prove that terminal diagnoses can eventually be transformed into manageable conditions through data-driven medicine.
