Artificial intelligence is making real progress in cancer detection, treatment planning, and drug discovery, but the sweeping promises from tech executives to "cure cancer" in the next decade significantly overstate what the technology can achieve. Cancer experts say AI will improve efficiency and personalization in care rather than deliver a universal cure, since cancer is not one disease but over 100 distinct conditions with different biology and treatment needs.

Why Can't AI Simply Cure Cancer?

The fundamental challenge is that cancer is not a single enemy. It's a catch-all term describing uncontrollable growth of abnormal cells, and the disease manifests differently depending on where it originates in the body. Brain tumors alone number over 100 distinct types, far more than in any other single organ. Each cancer type involves different genetic mutations, affects different cell types, and responds to different treatments. This complexity makes a universal AI cure biologically implausible.

Sherene Loi, a medical oncologist specializing in breast cancer at the Peter MacCallum Cancer Centre in Melbourne, explained the gap between hype and reality. "Until AI can actually think and do experiments itself at scale, I'm not sure that we're going to find a cure for all cancer or medical illness in the next 10 years," she stated. However, Loi and other cancer experts believe AI will deliver meaningful improvements in treatment efficiency, diagnostics, and pathology services that will enable more personalized and equitable cancer care.

Where Is AI Actually Making a Difference Today?

Despite the limitations, AI is already proving valuable in specific, measurable ways. Medical imaging stands out as a particularly promising area. Machine learning models trained on massive datasets of CT scans, X-rays, and tissue samples can detect patterns and changes that may be invisible to the human eye. Ajit Goenka, a radiologist and nuclear medicine specialist at the Mayo Clinic in Minnesota, has been exploring early detection of pancreatic cancer using machine learning since 2021.

Goenka's team discovered something striking: some patients who later developed pancreatic cancer had undergone CT scans months or years before diagnosis, but those scans were initially interpreted as normal. The researchers hypothesized that microscopic, measurable changes might exist that radiologists couldn't spot. In a proof-of-concept study published in 2022, machine learning tools successfully detected pancreatic tumors, including in pancreases that appeared normal to human radiologists, before clinical diagnosis. Further studies explored how robust and reproducible the method is across different institutions.

Felix Sahm, a neuropathologist at the University Hospital Heidelberg, leads a European collaboration called EUcanAI, which uses agentic AI (AI systems that can plan and execute multi-step tasks) to improve how brain tumors and central nervous system cancers are treated. The approach ensures that each step of a patient's journey informs the next one, making the process more efficient and less burdensome. In one example, a teenager with a brain stem tumor underwent surgery, biopsy, and rapid genetic sequencing to identify the mutations driving his disease. Traditionally, these results would go to a specialist board for discussion, taking additional time. Today, results can be delivered during surgery itself, saving patients time and money.

How AI Is Reshaping the Clinical Workflow

• Early Detection: Machine learning models can identify early signs of cancer in medical imaging that appear normal to human radiologists, potentially catching disease before symptoms emerge.
• Treatment Planning: AI systems can analyze tumor genetics and patient data to recommend personalized treatment strategies, moving away from one-size-fits-all approaches.
• Workflow Acceleration: Agentic AI can streamline the journey from diagnosis to treatment decision, reducing delays and improving patient outcomes by delivering results during procedures rather than after.
• Pathology and Diagnostics: AI tools assist in analyzing tissue samples and imaging, freeing up medical professionals from routine tasks so they can focus on complex clinical decisions.

The efficiency gains matter enormously in practice. Radiologists, pathologists, and oncologists spend significant time on routine image analysis and data review. AI can handle these tasks faster and sometimes more accurately, allowing clinicians to concentrate on the nuanced, human-centered aspects of care that machines cannot replicate.

What About the Broader Healthcare System?

Beyond cancer, AI is being deployed to address critical illness more broadly. Vanderbilt Health has been awarded a federal contract worth up to $11.2 million over five years to establish a platform for testing and validating AI-driven technologies aimed at improving outcomes for patients in intensive care units (ICUs). The initiative, called the Goldilocks Program, is part of a larger federal effort called CIRCLE, the Critical Illness Immunological Reprogramming and Control Point Learning Engine.

The problem CIRCLE addresses is urgent: more than 4.6 million people in the United States are treated in hospital ICUs each year for critical illnesses and injuries that trigger life-threatening immune and inflammatory responses. Clinicians currently cannot accurately track these responses, limiting their ability to intervene quickly and precisely enough to prevent organ damage or death. CIRCLE aims to harness advances in diagnostics and AI to speed development of next-generation interventions that identify and respond to harmful immune system dysregulation before irreversible organ injury occurs.

"CIRCLE represents an unprecedented opportunity to fundamentally change how we understand, monitor and treat critical illness," said Wesley Self, MD, MPH, Senior Vice President for Clinical Research at Vanderbilt Health and project lead for the award.

Wesley Self, MD, MPH, Senior Vice President for Clinical Research at Vanderbilt Health

The Goldilocks Program will test and validate new technologies across a network of up to 25 participating institutions and more than 100 intensive care units nationwide. Currently, it takes 12 to 18 months to move new health technologies from the readiness stage to first-in-patient testing. Goldilocks is designed to reduce that timeline to fewer than 90 days by providing continuous feedback, testing, validation, and clinical guidance throughout development.

Self emphasized the broader vision: "Critical illness remains one of the most complex and deadly challenges in medicine. This award positions Vanderbilt and our national partners to help create a new era of precision critical care that is smarter, faster, and more effective for patients and families".

Self

The Gap Between Promise and Progress

The contrast between tech executives' bold proclamations and what clinicians and researchers actually expect reveals an important truth about AI in healthcare. The technology is genuinely transformative in specific, bounded applications: detecting subtle patterns in medical images, accelerating diagnostic workflows, and identifying patients at highest risk. But these improvements, while significant, are not the same as discovering a cure.

The most honest assessment comes from the experts working at the frontlines of cancer care and critical illness. They see AI not as a silver bullet but as a powerful tool that can make medicine faster, more precise, and more equitable. That's a more modest promise than curing cancer in five years, but it's also one grounded in evidence and achievable in the near term.

" }