The rapid development of technology has led to the rapid advancement of cancer diagnosis and treatment methods. At the recent Berlin Science Week event, the Max Delbr ü ck Center for Molecular Medicine (MDC) showcased to the public an online tool based on artificial intelligence (AI) that can help doctors and patients find the most suitable diagnosis and treatment methods. Researchers believe that AI driven precision medicine in the future will provide better services for humanity. Candidate therapies often bring difficulties in selection. In the past decade, the number of diagnostic tests and available treatment methods for cancer has skyrocketed, with dozens of new cancer therapies approved each year, many of which are targeted therapies or immunotherapies based on precision medicine principles. Although new drugs and therapies are good news for patients, the increasing number of treatment options also means new challenges for attending physicians, namely how to choose the best treatment plan for patients. Dr. Altona Akalin, the leader of the "Bioinformatics and omics Data Science" group at MDC, sees opportunities in the increasingly complex process of cancer diagnostic testing and new therapies. He introduced at the Berlin Science Week event that "developing drugs and diagnostic procedures is a significant scientific task, but often takes decades to become useful products. We have developed a tool that can help clinical doctors make the best decisions based on patient conditions. We can also show people related to patients what other options they have so that they can make better decisions." The Akalin team has developed an AI based online tool, Onconaut. By entering simple keywords for search, such as "KRAS and lung cancer" (KRAS is an important site for genetic mutations in lung cancer), users can obtain a list of clinical studies in seconds. The list will display the latest clinical guidelines, a list of available drugs for KRAS mutant cancer, statistical data on related risks and treatment outcomes, etc. Akalin explained, "So far, its performance has been better than Google Doctor." AI is designed to assist, rather than replace, doctors. Akalin's team has provided training for Onconaut on various topics, including medical research and clinical guidelines published by official organizations such as the German Cancer Society and the American Society of Clinical Oncology, as well as data on difficult medical cases published in the New England Journal of Medicine. In order to further test and improve the tool, Akalin also collaborated with the Berlin Sharrit Medical School to train the model using real cancer patient data. Akalin emphasized that this tool can accelerate decision-making speed and improve the efficiency of experts. But it will never replace doctors. Similarly, researchers at the Berlin Schalke Medical School have also come to such a conclusion. They specifically studied the opportunities and limitations of large language models such as ChatGPT in automatically reviewing scientific literature to select personalized treatments, and concluded that AI can recognize personalized treatment plans in principle, but it is not yet close to the ability of human experts. Onconaut is one of the increasingly popular AI assisted precision medicine tools today. The precise treatment of cancer relies on the use of certain drugs, such as small molecule inhibitors or antibodies, to shut down overactive oncogenes. Developing personalized targeted drug treatment plans for patients typically requires a detailed understanding of the patient's tumor characteristics, such as phenotype, genetics, and tumor microenvironment, based on genetic testing results or the use of appropriate immunotherapy. Although existing treatment guidelines provide clinical decision support for doctors, typically only about 50% of patients meet the criteria of the treatment guidelines. Analyzing which therapies are expected to achieve the best results based on the patient's condition is a very complex process that often requires the integration of knowledge from various medical fields, including pathology, molecular pathology, oncology, human genetics, and bioinformatics. This is precisely where AI tools can play a role. AI tools can collect a large amount of daily treatment data from cancer patients and evaluate them using systematic research methods. Professor Uwe Ouler, the leader of the "Gene Regulation Bioinformatics" group that promotes precision medicine towards practical MDC, said, "AI can accelerate basic research, big data analysis, and even the exploration of treatment methods." AI is increasingly becoming an important component of cancer precision medicine in Germany. By focusing on molecular analysis, dynamic modeling, and advanced imaging techniques, AI is paving the way for more personalized and effective cancer treatments. AI excels in managing and analyzing common high-dimensional datasets in cancer research, including genomic, proteomic, and clinical medical data, and can extract relevant features from complex datasets to improve the accuracy of cancer diagnosis and the effectiveness of treatment strategies. AI algorithms can also predict the impact of genetic changes on protein structure and function, enabling medical experts to tailor treatment plans for patients based on their unique molecular characteristics of cancer. However, there are still some challenges in integrating AI into precision medicine. For example, ensuring the quality and representativeness of data, addressing ethical issues related to the use of AI, and how to integrate AI tools into existing healthcare systems. In this process, close collaboration between research institutions, healthcare systems, and technology companies is crucial for advancing AI driven precision medicine. (New Society)