The National Cancer Institute (NCI) is the U.S. federal government's principal agency for cancer research and training. As part of the National Institutes of Health (NIH), NCI coordinates a massive national research program, encompassing basic discovery science, clinical trials, epidemiology, and health services research. It directly conducts intramural research at its Bethesda campus and funds thousands of extramural grants, cooperative groups, and designated cancer centers nationwide. NCI's mission is to lead, conduct, and support cancer research across the spectrum to advance scientific knowledge and reduce the burden of cancer.

Why AI matters at this scale

With an annual budget exceeding $6 billion and a workforce in the thousands, NCI manages one of the world's largest and most complex cancer research portfolios. The institute sits atop a data deluge from genomics, medical imaging, electronic health records, and scientific literature. Traditional analysis methods are insufficient to synthesize this information into actionable knowledge. AI and machine learning are not just efficiency tools but foundational technologies for the next era of cancer discovery. They enable NCI to fulfill its mission at scale by extracting insights from data too vast or complex for human researchers alone, ultimately accelerating the translation of research into life-saving prevention strategies, diagnostics, and treatments.

Concrete AI opportunities with ROI

1. Accelerating Therapeutic Discovery: AI can model protein-drug interactions and screen virtual compound libraries against cancer targets, drastically reducing the time and cost of early-stage drug discovery. ROI is framed in years saved per investigational new drug and increased probability of clinical success.

2. Enhancing Cancer Surveillance: Machine learning models applied to national registry data (SEER) and real-world evidence can detect emerging cancer trends, disparities, and outcomes in near real-time. ROI manifests as more agile public health responses, better-targeted interventions, and improved population health metrics.

3. Optimizing Grant Review: Natural Language Processing (NLP) can triage and pre-review thousands of grant applications, matching them to ideal peer reviewers and flagging potential overlap. ROI is measured in reduced administrative burden on world-class scientists, faster funding cycles, and increased fairness in the review process.

Deployment risks for a large public entity

As a large government agency, NCI faces unique deployment risks. Procurement and Bureaucracy: Acquiring cutting-edge AI software or cloud compute can be slowed by federal acquisition regulations (FAR) and lengthy approval chains. Talent Retention: Competing with private sector salaries for top AI/ML talent is a persistent challenge. Interoperability and Legacy Systems: Integrating new AI tools with decades-old, mission-critical IT systems (e.g., grant management) requires significant customization and security review. Algorithmic Bias and Equity: Any AI tool deployed for research or public health must be rigorously validated across diverse populations to avoid perpetuating health disparities, requiring extensive upfront investment in bias testing and mitigation. Data Sovereignty and Security: Using sensitive patient data for AI training often necessitates expensive, secure on-premises or government-cloud infrastructure, rather than more agile commercial clouds.