Google’s new artificial intelligence (AI) Tool has cracked a problem in which it took scientists a decade to solve in just two days.
Jose pendes And at Imperial College London, his colleagues spent 10 years of how some superbugs get resistance to antibiotics – there is a growing danger Gives millions of people every year,
But when the team gave Google’s “co-scientist”-an AI Tool designed to collaborate with researchers-this question in a small signal, the AI’s response produced the same answer in just two days as their then unpublished conclusions.
Surprisingly, the Pendes emailed Google to check if they had access to their research. The company replied that it was not. Researchers published their findings at prepromt server on 19 February BiorxivSo they have not yet been reviewed by the colleague.
“Our findings show that AI has the ability to synthesize all available evidences and direct us to the most important questions and experimental designs,” Tiago dias da costaA lecturer in bacterial pathogenesis at Imperial College London, Said in a statement“If the system works and at the same time we hope that it can be, it can be a game-changing; to exclude the ‘dead end’ and effectively enable us to progress at an extraordinary speed.”
Using AI to fight superbugs
Anti -antimicrobial resistance (AMR) occurs when infectious microbes – such as bacteria, viruses, fungi and parasitic – obtain resistance to antibiotics, provide ineffective to essential drugs. was dubbed “Mute epidemic“AMR represents one The biggest health threat Facing humanity as Excessive use and misuse of antibiotics Its prevalence increases in both medical and agriculture.
According to a 2019 report Disease Control and Prevention Center (CDC)Drug-resistant bacteria killed at least 1.27 million people globally that year. About 35,000 of those deaths were in the US alone, which means that the issue had increased by US confusion by 52%. CDC’s last AMR Report, 2013,
To examine the problem, the pendes and their team discovered a type of superbug methods, a family of the bacteria-enacted virus, known as the capsid-forming phase-inductible chromosomal island (CF-Pickis)-Acquisite their capacity to infect the species of bacteria.
Connected: Dangerous ‘superbugs’ are a growing threats, and antibiotics cannot stop their rise. What can happen?
Scientists envisaged that these virus did this with a tail, which is used to inject viral genomes in the host bacterial cell, from various bacterial-enacted viruses. Experiments proved their hump to correct, revealing a success mechanism in horizontal gene transfer that the scientific community was already unaware.
Before anyone in the team publicly shared their findings, researchers presented the same question for Google’s AI co-scientific equipment. After two days, AI returned the suggestions, being the one he knew as the correct answer.
“It effectively meant that algorithm was able to look at the available evidence, analyze possibilities, look at the design experiments, and propose a lot of hypothesis, which we had arrived in the years of labor research, but in a fraction of that time,” Pendes, a Professor of Microbiology in London, said that the statement said.
Researchers stated that using AI from the beginning would not have removed the need to conduct experiments, but this would have helped them come up with hypothesis, thus helping them save years of work.
Despite these promising conclusions and OtherThe use of AI in science remains controversial. For example, A-assisted research has been an increasing body Shown to be inaccessible Or even lump sum FraudScientists are proposed to reduce these problems and maximize AI for research. Equipment to detect AI malpractice And establishing a moral outline to assess the accuracy of findings.
