The intriguing story that went viral online gained significant global attention in March 2026, demonstrating how a determined individual combined his data expertise with cutting-edge AI tools to achieve what was previously considered impossible for a private citizen. Paul Conyngham, an Australian AI and data analysis expert, refused to accept a terminal diagnosis for his rescue dog, Rosie, who had aggressive mast cell cancer and only months to live despite traditional chemotherapy. The owner said tearfully, “She’s my best friend. Now it’s my turn to fight her.”
Paul spent $3000 (AUD) to have Rossie’s healthy and tumor DNA sequenced at the UNSW Ramaciotti Centre for Genomics, which produced approximately 300GB of genetic data. He used ChatGPT as a research assistant to formulate a “plan of attack”, research immunotherapy, and draft the complex 100-page ethics application required to administer an experimental drug in Australia.
Paul used Google DeepMind’s AlphaFold to model the 3D structure of Rosie’s mutated proteins. He compared healthy and tumor DNA to pinpoint mutations and used his own machine-learning algorithms to identify seven negotiations that would most likely trigger an immune response. Subsequently, he collaborated with the UNSW RNA Institute to synthesize a personalized mRNA vaccine based on his data-the first time such a vaccine was designed for a dog by a non-scientist. The vaccine led to a significant reduction in Rosie’s primary tumor; her mobility and energy levels returned, allowing her to run and play again.
While not a complete cure, the treatment has transformed a terminal sentence into a manageable condition. Paul is currently working on a second-round vaccine to target remaining tumor cells that were resistant to the first dose.
Ultimately, this discovery will pave the way for democratizing personalized medicine, potentially making high-end cancer treatments more accessible and faster to develop for both animals and humans.