How an AI solved in two days a problem about superbacteria that took scientists years

Professor José R. Penadés and his team of Imperial College in London have spent years working and demonstrating why some superbacteria are immune to antibiotics.

Now, Penadés gave “cooking” – an artificial intelligence system created by Google – a brief question about the central problem he had been investigating and The tool reached the same conclusion in 48 hours.

The scientist told the BBC his surprise when he discovered what he had done, since his research had not been published, so the AI ​​system could not have found it in the public domain.

“I was shopping with someone and I said: ‘Please leave me only one hour, I need to digest this,” he told the BBC Radio 4.

“I wrote an email to Google to say: ‘You have access to my computer, isn’t it?'” He added.

The technological giant denied having that access.

The full decade that scientists also used to include the time that took them to validate the research, which was several years.

But they say that If they had had the hypothesis at the beginning of the project, years of work would have been saved.

Professor Penadés said the tool had done more than successfully reproducing his research.

“It’s not just that the main hypothesis it offers was correct,” he said.

“It is that it offers four and all made sense.

“And in the case of one of them, we had never thought about it, and now we are working on it”He said.

Playing of superbacteria

Researchers have been trying to find out how some superbacteria are created (dangerous germs that are resistant to antibiotics).

Their hypothesis is that superbacteria can form a tail from different viruses that allows them to spread between species.

The Penadés Professor illustrated him by ensuring that it is as if the superbacteria had a “key” that allowed them to move from home at home, or from a host species to another.

Fundamentally, this hypothesis was exclusive to the research team and had not been published anywhere else. No team member had shared his findings.

So Penadés was happy to take advantage of this to test Google’s new artificial intelligence tool.

Only two days later, artificial intelligence returned some hypotheses, and its first thought, the main response provided, suggested that superbacteria could form queues exactly in the manner described in their research.

“This will change science”

There is a lot of debate for the impact of AI.

Its defenders say it will allow scientific advances, while others fear that you eliminate jobs.

Penadés Professor said that He understood why the fears of the impact on jobs like his were the “first reaction” of the people, but added that “when you think about it, it is rather that there is an extremely powerful tool.”

He said the project researchers were convinced that it would be very useful in the future.

“I think this will change science, no doubt,” said Penadés.

“I am facing something spectacular and I am very happy to be part of that.

“It’s like having the opportunity to play a great game; I feel that I am finally playing a game of the Champions League with this thing,” he said.