AI Today Podcast: AI in Pharma – Interview with Xiong (Sean) Liu, Novartis
Dec 8, 2023
auto_awesome
Xiong (Sean) Liu, Director at Novartis, discusses the impact of AI in pharma including streamlining processes, optimizing drug discovery, and enhancing patient care. He shares real-world examples of AI applications in biology discovery, molecule design, and clinical trials. The chapter also emphasizes the transformative impact of AI and the importance of ethical AI in the pharmaceutical industry.
AI has the potential to revolutionize molecule discovery and improve clinical trials in the pharmaceutical industry.
Addressing challenges like performance, validation, and ethical concerns is crucial for successful AI integration in highly regulated industries.
Deep dives
The Importance of AI Adoption
AI adoption is crucial across industries, including highly regulated ones like pharmaceuticals. The power and potential of AI make it a must-have technology for organizations. By leveraging AI, companies can improve efficiency, drive innovation, and enhance decision-making processes. However, it is essential to have the right execution plan that aligns with business needs. Additionally, a focus on ethical and trustworthy AI is necessary to ensure responsible and beneficial use of the technology.
Opportunities in AI for Pharmaceuticals
AI offers numerous promising opportunities in the pharmaceutical industry. In the realm of early biology discovery, the application of large language models and foundation models on biological language data shows great potential for in silico experiments and gene-cell annotation. Generative AI methods can revolutionize molecule discovery by designing new molecules and reducing the screening task. AI has the potential to improve clinical trials through protocol writing, clinical site selection, patient enrollment, and the application of real-world data for external control arms. These applications already show promising results and are expected to continue evolving and expanding.
Challenges in AI Adoption for Highly Regulated Industries
AI adoption in highly regulated industries like pharmaceuticals comes with its unique challenges. Performance is one of the key concerns, as AI methods may not always deliver perfect results. Achieving higher performance levels, ideally above 90%, is crucial for gaining trust and widespread adoption. Validation is another challenge, especially when AI models make predictions without experimental evidence, creating uncertainties in the research domain. Finally, concerns related to regulation, risk, ethics, and the fear of AI replacing humans can impact the adoption of AI solutions. Addressing these challenges is necessary to foster successful AI integration in highly regulated industries.
The Future of AI in Pharmaceuticals and Healthcare
The future of AI in the pharmaceutical and healthcare sectors looks promising. AI models are expected to become more robust and capable of generalizing across experimental data, leading to improved performance and wider adoption. Model pipelines that connect various fundamental models and workflows have the potential to transform efficiency and accelerate discoveries. However, it is crucial to always prioritize the core value of healthcare: bettering the lives of patients. As AI continues to advance, regulations, rules, and additional guidance are expected to emerge, promoting trustworthy and ethical AI practices.
AI is impacting very about every single industry and pharma is no exception. AI has tremendous potential in streamlining processes, optimizing drug discovery, and enhancing patient care in ways we’ve never seen before. In this episode of the AI Today podcast we have an insightful conversation with Xiong (Sean) Liu. He is the Director of Data Science and AI at Novartis.