A Panel Discussion titled "The Impact of AI in Drug Discovery: Present and Future"

I took part in the 2024 CDD Korean Life Science Community Meeting on July 18, 2024. The event included excellent research presentations from speakers representing several Korean companies, as well as informative introductions to the features of the CDD Vault system. I, as the Chair, moderated a panel discussion titled “The Impact of AI in drug discovery: present and future” featuring three panelists: Dr. Yoo-Young Song from Standigm, CEO Art Cho from InCerebro, and CEO Jae-Moon Choi from Calici. This event was specifically for Koreans, so all sessions, including the panel discussion I led, were conducted in Korean, except for foreign speakers.

I sent the questions to the panelists in advance and had a Zoom meeting, but the discussion exceeded my expectations in terms of engagement.

Dr. Kyung-Seok Oh from Daewoong Pharmaceutical asked me a challenging question before the panel discussion: “Does AI truly contribute to drug discovery? If yes, how much?” While this question brought up the panel discussion’s theme again, it also introduced a quantitative element that could spark varied opinions among individuals and groups. While this could have been the last question of the panel, I believed it would be beneficial to seek the perspectives of various experts in attendance. So, I sought the thoughts of the audience.

Dr. Jin-Geon Bae, a prominent figure in Korea’s pharmaceutical industry, revealed how AI technology decreased the time frame for his project from 3 years to just 3 months. CEO Taegon Baik from Arum Therapeutics and Dr. Wooseok Han from Cyrus Therapeutics provided highly realistic perspectives as well. The comprehensive recounting of these stories prior to the discussion greatly improved the overall effectiveness of the conversation.

Here are seven questions that I had prepared ahead of time.

1. Could you give a brief introduction to yourself and your company?

2. There are two major trends in AI drug discovery: one uses AI technology to improve tasks in the drug discovery process (task improvement), and the other uses AI technology to innovate the drug discovery process itself (innovate workflow development). Which option do you and your company lean towards?

3. If you had ample resources, which sector would you prioritize for investment?

4. How do you assess the success and failure of collaborative or in-house projects?

5. Which market should we concentrate on first: global or Korean?

6. Which modality do you think AI technology can be better applied to? On which one are you placing your focus?

7. If a company or individual is interested in adopting AI technology, what advice would you provide?

The panel discussion didn’t provide complete answers to all questions, but it did facilitate the exploration of commonalities and differing opinions. Many companies in the AI drug discovery sector are prioritizing small successes over disruptive research, given the current financial situation. They prioritize reinforcing their core expertise rather than acquiring new abilities. In addition, the absence of effective communication is a shared challenge in all projects. Whether it is collaboration or in-house projects, communication based on mutual respect among researchers with different backgrounds is always crucial.

As expected, there were differing views as well. Determining whether it’s best to prioritize the global market or establish a strong position in the Korean market first is a tough decision. Depending on their business models and market conditions, companies make a range of different choices. There were varying opinions on the most suitable modality for applying AI technology, with certain companies concentrating on conventional small molecule drug discovery while others remained interested in emerging modalities like ADC and DAC.

Precisely defining each term and narrowing the topic is advisable for in-depth discussions. However, just thinking about the present and future centered on two large terms–AI and drug discovery–provided some insights for all panelists and audiences. Given that all the panelists specialized in AI drug discovery, their viewpoints were largely aligned. By presenting a subject that underscores their distinctions, it would encourage more comprehensive discussions. A panel discussion featuring drug discovery experts from different disciplines may expose key areas of disagreement, creating an intriguing event.

Finally, I want to express my sincere appreciation to Collaborative Drug Discovery Inc. for their support in organizing such an engaging panel discussion, and I would also like to thank the panelists and attendees for their active involvement.

Key Takeaways:

1. AI’s impact on drug discovery: The discussion highlighted AI’s potential in significantly reducing drug discovery time, as evidenced by Dr. Jin-Geon Bae’s experience of reducing a process from 3 years to 3 months.

2. Communication is Crucial: Efficient communication based on mutual respect among researchers with diverse backgrounds is essential for successful collaboration in AI-driven drug discovery projects.

3. Market Focus: There were varied strategies on whether to prioritize global markets or establish a strong domestic presence first, reflecting diverse business models and market conditions among companies.