Artificial Intelligence is a Bridge Across Expertise

Paul Conyngham, an Australian AI consultant with no formal medical or scientific training, turned to generative AI after his dog Rosie’s terminal cancer failed to respond to conventional treatments. Rosie had undergone multiple surgeries, chemotherapy, and immunotherapy, with vets estimating her remaining lifespan at one to six months. Conyngham sequenced Rosie’s tumor and used ChatGPT alongside AlphaFold to design neoantigens, collaborating with researchers to develop a personalized mRNA cancer vaccine. The vaccine, combined with an immune checkpoint inhibitor, was administered to Rosie. While not a cure, it led to a significant reduction in her largest tumors and improved mobility. The process relied on AI tools to bridge expertise gaps, but critical steps—sequencing, lab work, and scientific validation—were executed by real institutions and specialists. Conyngham’s approach demonstrates how AI can serve as a bridge across domains, enabling non-experts to coordinate complex scientific workflows. The project required rapid decision-making, ethical oversight, and logistical coordination, all accelerated by AI’s analytical support. Public narratives often overlook the behind-the-scenes efforts, including failed attempts, administrative hurdles, and time constraints, which were pivotal in Rosie’s case. The story underscores a broader potential: AI’s role isn’t replacing experts but augmenting human judgment with speed and precision. In high-stakes scenarios like terminal illness, this hybrid approach—combining machine intelligence with human persistence—can make a tangible difference. Conyngham’s work with Rosie highlights how AI tools may reshape personalized medicine by democratizing access to scientific collaboration.