image: ©dani3315 | iStock
Researchers at the University of South Florida are investigating how artificial intelligence could transform the future of cancer treatment, vaccine development and drug discovery
A new study published in Nature Machine Intelligence examines whether AI systems can reliably predict how the human immune system reacts to threats such as viruses, tumours, and harmful proteins.
The research, led by scientists at the University of South Florida, shows that while advanced AI tools are helping researchers move faster than ever before, these systems still require real-world testing before they can safely guide patient care.
AI and the immune system
The immune system relies on specialised cells to identify substances that do not belong in the body. These substances, known as antigens, can come from viruses, bacteria or cancer cells. Once detected, the immune system launches a targeted defence to destroy the threat.
Researchers focused on an AI model called PanPep, designed to predict how immune cells known as T cells recognise and bind to antigens. This interaction is one of the most important steps in determining whether the body can fight infections or respond to treatments such as immunotherapy.
The study introduced a new framework for evaluating how accurately AI models perform these predictions under realistic conditions rather than relying only on controlled laboratory data.
Why accuracy matters
Scientists believe AI could dramatically speed up the search for new therapies by narrowing down which drug or vaccine candidates are most likely to work. Instead of running thousands of expensive and time-consuming laboratory experiments, researchers could use computer models to identify the strongest possibilities first.
According to the study, tools like PanPep may eventually allow scientists to simulate parts of cancer screening and treatment development on computers, potentially reducing timelines from years to just days.
This could be especially important for patients with advanced cancers, where identifying an effective treatment quickly can make a major difference.
However, the researchers also found that AI systems can struggle when faced with entirely new or rare immune targets. In some cases, models may misinterpret signals or produce biased predictions, raising concerns about using them too early in clinical settings.
The team believes their framework could also be applied to broader immunology research, including studies of antigen presentation and other immune system interactions.
The future of AI in medicine
The findings are another clear step toward personalised medicine, where treatments are tailored to an individual’s immune system and disease profile. Although AI-driven healthcare tools are not yet ready to independently guide medical decisions, researchers say continued testing and refinement could eventually make them powerful partners in clinical care.
For now, the study serves as a reminder that while AI has enormous potential in medicine, human oversight and rigorous scientific validation remain essential before these technologies can be fully trusted in real-world healthcare settings.