What happens when a data scientist refuses to accept a terminal diagnosis for his dog? In this case, a world first.
When veterinarians told Sydney tech entrepreneur Paul Conyngham that his rescue dog Rosie had months to live, he didn't accept the prognosis. Instead, the data scientist and machine learning engineer — with 17 years in the field but no background in biology — turned to ChatGPT and set out to do something no one had done before: build a personalised cancer vaccine for a dog.
ChatGPT suggested immunotherapy and directed Conyngham to the University of New South Wales, where he paid for Rosie's genomic sequencing. He also used AlphaFold, an AI tool from Google's DeepMind, to find mutated proteins that could be potential targets for treatment. The resulting formula was handed to researchers at UNSW's RNA Institute, who manufactured the vaccine in under two months.
The vaccine pushed back approximately 75% of the cancer. By the end of January, Rosie was jumping over a fence to chase a rabbit — a far cry from the dog who had barely been able to move weeks earlier. It is important to note that this is a single case with no controlled trials, and Conyngham is already working on a second vaccine for a tumour that did not respond.
Researchers involved believe the approach could one day extend to human patients and beyond. A fascinating glimpse at what AI-assisted medicine may look like in the years ahead.
