The search giant teamed up with Imperial College London and its Fleming Initiative, handing scientists a flashy new AI—built with its Gemini 2.0 tech—to speed up research. The results, as expected, left researchers wondering if they’d been wasting the last decade.
Professor José Penadés and his team had spent ten years figuring out how superbugs develop antibiotic resistance. Then, on a whim, they let Google’s “co-scientist” AI take a crack at the question.
Within 48 hours, the AI spat out the same answer they’d spent a decade piecing together. Understandably paranoid, Penadés immediately emailed Google to check if it had somehow nicked his unpublished work. Google, playing the innocent tech overlord, assured him it hadn’t.
The research team was left amazed and slightly horrified. Co-author Tiago Dias da Costa gushed about AI’s ability to sift through mountains of evidence, cut through dead ends, and suggest the most relevant experiments—meaning scientists could theoretically get to the exciting bits without wasting years of their lives in the lab.
Professor Mary Ryan, Vice Provost (Research and Enterprise) and Imperial Global USA Academic Theme Lead for Advanced Materials and Cleantech, at Imperial College London, said: "The world is facing multiple complex challenges – from pandemics to environmental sustainability and food security. Addressing these urgent needs means accelerating traditional R&D processes, and artificial intelligence will increasingly support scientific discovery and pioneering developments.
“Our scientists are among the most talented in the world, with the curiosity and lateral thinking needed to exploit AI technologies for societal good. The prospects could be game-changing, starting with new avenues for biomedical research and sowing the seeds for greater scientific efficiency.
Google insists its AI isn’t here to steal boffin’s jobs, it just “collaborates” with researchers by answering questions, generating hypotheses, and proposing experiments.
Of course, not everyone is thrilled about AI-playing boffin. The field is already grappling with a flood of AI-generated research that’s impossible to replicate or just outright fraudulent. But that hasn’t stopped Google from publishing glowing test results of its AI co-scientist system.
