Recent AI advancements in biotech are less about new algorithms and more about reaching a critical threshold of complete, high-quality data from electronic health records. This allows AI to extract genuine insights rather than just compensating for historical data shortcomings.
Many promising biomarkers are merely correlated with a disease, not a cause. Developing effective drugs requires targeting mechanisms with a direct causal effect on the outcome. This distinction is so critical that Valo Health has a "Chief Causal AI Officer" role to emphasize the focus.
Beyond analyzing existing datasets, a significant benefit of AI is its ability to create higher-quality data in the first place. For instance, accurate, automated transcription of doctors' notes improves the richness and reliability of clinical data, creating a virtuous cycle for future analysis.
By modeling a complete patient trajectory using multi-omics data, AI could predict how a patient would fare on a placebo or standard care. This would allow for "synthetic" control arms, reducing the ethical and practical challenges of enrolling patients in non-treatment groups.
While there's ample evidence from lab experiments (causal, but not human) and patient data (human, but not causal), the most promising drug targets are found where these two overlap. This "human causal evidence" is rare and difficult to obtain, but provides the strongest signal for success.