LIFESCIENCE & AI @SF Permit Center (Room 134)
Wednesday January 14th, 2026
Life Sciences and AI:
A Mathematical and Evidence-Based Perspective on Realistic Expectations
In the era of rapid AI advancement, tools like AlphaFold have revolutionized structural biology, enabling unprecedented predictions of protein structures and interactions that accelerate target identification and drug design. Yet, AI is not a panacea for the inherent complexities of biology—clinical success rates remain challenging, with high attrition in later phases despite early promises.
This investor-only session on the morning of January 14 (during JPM Healthcare Week 2026) offers a balanced, data-driven examination of AI's true value in life sciences. CEO Axel Tillmann will open with an introduction, followed by one or two expert-led panels discussing mathematical limitations (e.g., probabilistic modeling vs. biological irreducibility), proven accelerations (e.g., Phase I success rates of 80-90% for some AI-designed molecules vs. historical 40-65%), and critical due diligence factors for evaluating AI-enabled companies.
Investors will gain insights into distinguishing genuine breakthroughs from hype, essential for robust portfolio decisions amid ongoing clinical readouts and regulatory evolution.
Discussion Details:
Hype vs. Clinical Reality: Despite promises, ~90% clinical failure rates persist. Many AI-designed candidates fail in later phases or show non-significant efficacy (e.g., Insilico's Phase IIa shortfall in 2024-2025). Partnerships since 2012 yielded few Phase II advances.
Hallucinations and Reliability Issues: Generative models produce plausible but incorrect outputs (e.g., fabricated structures or interactions). In biology, this risks misleading designs; models extrapolate poorly beyond training data.
Mathematical and Data Limitations: Biology involves computational irreducibility—systems too complex for full prediction. Biases, incomplete data, and black-box nature reduce interpretability. AlphaFold excels at folding but less at dynamics or off-distribution predictions.
Regulatory, Ethical, and Adoption Challenges: Data privacy, bias, explainability hurdles; smaller firms struggle with implementation. Biosecurity risks (e.g., misuse for harmful designs) noted.
Event Details
Date: January 14, 2026
Time: 10:00 AM - 11:30 AM
Location: SF Permit Center
Format: In-person
Spaces are limited—secure your spot today!
Why Attend?
Gain actionable frameworks to streamline due diligence and enhance decision-making.
Network with industry leaders, investors, and experts.
Walk away with a renewed knowledge of underlying mat