ESMO Issues First Guidelines for Safe Use of AI Language Models in Oncology

Oct 20, 2025
The European Society for Medical Oncology has released its first framework for the responsible use of large language models in cancer care, outlining safety and governance standards for patient, clinician, and institutional applications.

The European Society for Medical Oncology (ESMO) has published its first guidance for the safe use of artificial intelligence language models in oncology, announced in a press release. The framework, called ELCAP (Guidance on the Use of Large Language Models in Clinical Practice), was introduced at the ESMO Congress in Berlin and published in the Annals of Oncology.

ELCAP establishes standards for integrating large language models into cancer care while ensuring patient safety and clinical oversight. It categorizes AI applications into three groups: patient-facing tools like chatbots that must operate under supervision, clinician-facing tools that require validation and accountability, and background systems used in hospital infrastructure that demand institutional monitoring.

ESMO stated that the guidance aims to protect patient data, support clinicians, and promote trustworthy AI use in clinical environments. With over 45,000 oncology professionals in its network, the organization described ELCAP as a step toward making AI a reliable and safe component of oncology practice.

We hope you enjoyed this article

Free newsletter

Life AI Weekly

Weekly coverage of AI applications in healthcare, drug development, biotechnology research, and genomics breakthroughs.

Whitepaper

Tensordyne Napier: What If One Rack Could Do the Work of Nine?

Tensordyne

This Tensordyne whitepaper presents Napier, an inference-focused AI processor and rack-scale system based on the company’s TDN Math logarithmic number system. It examines infrastructure requirements for large mixture-of-experts and agentic models, compares major inference architecture approaches, and details the TDN AIP processor, TDN72 pod, TDN Link fabric, and Napier Ultra configuration. The paper reports simulation-based performance, cost, and accuracy-validation results, including Tensordyne’s projected comparison of one Napier rack with a nine-rack Nvidia Rubin plus Groq deployment; the chip is reported as taped out and in fabrication.

Read more
Free, six days a week

Daily AI Brief: the AI news that matters, in your inbox.