AI Solves Superbug Mystery in Just Two Days—A Task That Took Scientists Years

In partnership with

Learn AI in 5 minutes a day

This is the easiest way for a busy person wanting to learn AI in as little time as possible:

1. Sign up for The Rundown AI newsletter

2. They send you 5-minute email updates on the latest AI news and how to use it

3. You learn how to become 2x more productive by leveraging AI

Sign up to start learning.

A scientific challenge that took microbiologists a decade to unravel has been cracked in just two days by an advanced artificial intelligence (AI) tool.

Professor José R. Penadés and his team at Imperial College London had spent years investigating why certain superbugs develop resistance to antibiotics. However, when they tested Google’s AI tool, known as "co-scientist," by providing it with a short prompt related to their research, the AI reached the same conclusion within 48 hours.

Prof. Penadés expressed his astonishment, noting that his findings had never been published, meaning the AI could not have sourced them from publicly available data.

"I was out shopping when I realized what had happened. I told my companion, ‘Give me an hour, I need to process this,’" he shared with BBC Radio Four’s Today program. Concerned about how the AI had accessed his research, he even reached out to Google, which confirmed it had not accessed his computer.

The decade-long research process also included years spent proving the findings. However, had the researchers started with the AI-generated hypothesis, it would have saved them significant time.

AI's Potential Beyond Replicating Research

Beyond simply replicating the team’s findings, the AI tool provided four additional hypotheses—all of which made scientific sense. One of these ideas had never occurred to the team, prompting them to explore it further.

The researchers had been investigating how superbugs—antibiotic-resistant bacteria—emerge. Their hypothesis suggested that superbugs might acquire virus-like tails, allowing them to spread between different species. Prof. Penadés described it as bacteria gaining "keys" that enable them to move from one host to another.

Since this hypothesis was unique and unpublished, it was the perfect test case for Google’s AI. Incredibly, the AI returned with a top suggestion that mirrored the researchers' own findings.

"This Will Change Science"

The growing role of AI in research is a topic of debate—some hail its ability to accelerate discoveries, while others fear it could replace human jobs.

Prof. Penadés acknowledged these concerns but argued that AI is ultimately an "extremely powerful tool" that can aid researchers rather than replace them.

"I believe this will transform science," he stated. "It’s an incredible opportunity—like playing in the Champions League of scientific discovery."