LSX Nordic Congress’ Post

Thank you LSX - partnering for Life Science eXecutives for inviting me to discuss the hottest topic: when and how AI will impact life science. The take-home? When? - Now. All major pharma companies already have significant efforts in using AI to improve drug development. As Julie Klint shared nuances about, AI has proven itself useful in small molecule design, yielding sizable improvements in speed and quality. - Soon. There are several interesting readouts of AI-generated assets (depending on your definition of AI-generated) in the coming 18 months, which can validate some of the AI-powered biotechs, which so far have resulted in few approved assets. The proof is in the pudding. How? - Two main ways: generic capabilities (e.g. reading and summarizing scientific literature, meeting summaries, internal coordination) at first and specific ones (e.g. target identification, lead generation) after that. - All pharma companies require some in-house knowledge to navigate and manage the implementation of AI, but most will stay true to their existing core business and not become software companies. Thus the need for external innovators (e.g. recent partnership between Iambic Therapeutics and Lundbeck) - Challenges and uncertainties remain in big pharma, e.g. in organizing data to be useful to AI models, changing processes and ways of working. That's why the first processes to be improved by AI will be those with fewer interdependencies, which can relatively be automated/improved without having to change many things at once - It seems as though those who have come far, e.g. Genmab as Hisham explained, may reap over-sized benefits as the models become more powerful and capabilities and use cases expand. Thank you for a great discussion Julie Klint, Hisham Hamadeh, Paul Ainsworth, Daniel Schmidt! #lsx #ai #lifescience #venturecapital