What changes in everyday life?
Better candidate selection could eventually shorten parts of the long journey to new treatments.
A researcher spends less time checking unpromising molecules and more time investigating candidates with a stronger rationale.
A researcher spends less time checking unpromising molecules and more time investigating candidates with a stronger rationale. artificial intelligence can already suggest and rank candidate molecules. The big step is connecting these suggestions to reliable experiments. Better candidate selection could eventually shorten parts of the long journey to new treatments. Drug developers could focus expensive laboratory work where evidence suggests the best chance of progress. Imagine removing thousands of weak possibilities from a drug search before spending money making them. Scientists could concentrate their efforts on the molecules with stronger reasons to succeed, potentially speeding up the earliest stages of discovery. But computers cannot tell us everything about how a medicine behaves inside a person; careful experiments and clinical trials still determine whether it is truly useful.
Better candidate selection could eventually shorten parts of the long journey to new treatments.
Drug developers could focus expensive laboratory work where evidence suggests the best chance of progress.
This is an opportunity we're exploring, not a promise that the complete result is already available. The research below explains the difference.
This idea comes from the Yottabit research foundation's Opportunity Atlas. It is a proposed application rather than a quantified forecast or evidence of a broadly deployed product.
What would demonstrate real progress? A useful result that works reliably outside a demonstration, withstands appropriate testing and improves an outcome people care about.