The future of cancer treatment is being shaped by the quiet, steady progress of artificial intelligence, driven by data, regulation, and the pursuit of precision and sustainability. As younger people are increasingly diagnosed with cancer, the need for kinder, more effective medicines becomes paramount. This demographic shift is one of the key forces driving the application of AI in oncology, with a focus on making treatments more precise and less harmful.
Sai Jasti, senior vice president and head of data science and AI at Bayer, highlights the rising incidence of cancer among younger patients, which necessitates longer-term therapies. This shift in demographics is prompting a reevaluation of drug design, with a focus on toxicity and long-term side effects. Jasti emphasizes the importance of developing kinder medicines, aligning with Bayer's oncology strategy for precision drug development.
Precision, in this context, does not imply bespoke drugs for every patient, but rather a gradual shift towards a deeper understanding of disease biology. Large, multimodal datasets, including molecular, cellular, clinical, and real-world evidence, are being utilized to detect patterns that would be challenging for humans to identify. This incremental approach is deliberate, as the adoption of new technologies in oncology progresses slowly.
Ofer Sharon, MD, CEO of OncoHost, acknowledges the slower pace of change in medicine compared to consumer AI. Clinical AI systems require extensive validation through trials to ensure safety, and regulatory pathways are designed for static products, not evolving models. As a result, progress in oncology has been more about integrating AI into existing clinical frameworks rather than headline-grabbing applications.
The most tangible advances have come from machine learning systems that integrate complex biological data, such as genomics, proteomics, imaging, and clinical records, to identify patterns that are difficult for clinicians to detect. These capabilities are gradually moving oncology towards more personalized care, but within strict constraints. Most cancer patients are still treated according to high-level clinical guidelines, prioritizing consistency over individual variation.
Sharon suggests that the near-term trend is towards increasing the resolution of this system, breaking broad categories into smaller subgroups defined by molecular, clinical, or protein-level characteristics. Biomarkers are emerging as the bridge between AI insights and routine clinical decision-making. To ensure AI's implementation in clinical practice, it needs to become mainstream.
Regulators are responding by requiring biomarkers alongside new drugs and streamlining approval pathways for diagnostic tools that support treatment decisions. At large pharmaceutical companies like Bayer, AI deployment across the oncology value chain is being influenced by these shifts, with the greatest gains seen in early-stage research. Bayer is developing an agentic AI framework for the Israeli market to improve access to medicines, aiming to bring it to market next year.
While the full potential of AI in oncology may take several years to realize, executives emphasize that the field is being driven forward by infrastructure improvements, better data, better models, and tighter integration between diagnostics, drug development, and clinical decision-making. As cancer increasingly affects younger populations, the pressure to make treatments effective and sustainable over the long term is likely to intensify.
