tech
Two AI-based science assistants succeed with drug-retargeting tasks
Both tools generate hypotheses; one goes on to analyze some of the data.
TL;DR
- Two AI systems, Co-Scientist (Google) and Robin (FutureHouse), aim to assist scientists with hypothesis generation and testing.
- These systems are designed to handle the overwhelming volume of scientific literature, a challenge for human researchers.
- Co-Scientist acts as a 'scientist in the loop,' where researchers guide the AI's hypothesis formulation and evaluation process.
- Robin can analyze biological data from specific experiments, drastically reducing the time required compared to human analysis.
- Both systems have shown success in identifying potential drug repurposing for conditions like leukemia and macular degeneration.
- While effective for certain tasks, these AI systems are not yet capable of solving more open-ended or complex scientific problems.
- The use of specialized tools within the AI systems is crucial for accuracy and preventing 'hallucinations' of false information.