Updated 12 sec ago · auto-refreshing
Trending claims
The fastest-moving factual assertions across the open web, ranked by velocity and weighted by credibility.
The AI-driven drug discovery work took less than a day and cost about $46 in computing power.
AI agents analyzed studies on the B7-H3 protein and recommended a successful targeting approach for lung cancer.
The language model used in the research had a knowledge cutoff of January 2025.
Merck did not respond to the author's request for comment.
Dr. Charles Rudin stated that no other drugs targeting the same mechanism have been tested in a clinical trial.
AI bots cannot currently perform certain parts of the drug discovery process, such as running lab experiments with cells.
The AI bots used in the research were not connected to the internet.
Dr. Zou expanded his AI team to thousands of agents organized into departments of a virtual biotech company.
Researchers created a leaderboard for AI agents, rewarding top performers with more computational resources.
Dr. Zou stated that a human expert always reviews the AI agents' work before proteins are synthesized in real life.
Researchers made AI agent personas more critical and added 'devil's advocate' bots to challenge ideas.
For the COVID-19 vaccine problem, the lead AI selected agent specialties in immunology, computational biology, and machine learning.
Kyle Swanson is a former graduate student in Dr. Zou’s lab and was involved in the Covid vaccine research.
Dr. Pak is a scientist who specializes in the type of research the AI bots were conducting.
A Stanford lab reported in a Science journal paper that AI agents predicted the most effective approach for attacking a lung cancer protein.
AI bots designed dozens of novel proteins for new Covid variants, and testing showed two of them worked.
A large pharmaceutical company (Merck) announced promising results for a drug using the same approach the AI agents independently identified.
John Pak and collaborators tasked AI bots with redesigning part of a Covid-19 vaccine in 2024.
Rafael Pinilla-Redondo stated he would be surprised if VIPR is the last unexpected genetic targeting system to be discovered.
Dr. Yoon stated that VIPR is like a 'gun that viruses are pointing at each other,' which, if stolen by bacteria, becomes an immune system.