For more crisp and insightful business and economic news, subscribe to The Daily Upside newsletter. It's completely free and we guarantee you'll learn something new every day.

The devil -- and the doctor -- is in the details.

Google's cloud division said Monday that it's launching a new AI-powered tool designed to rifle through patients' medical records and pull them together into a neat bundle, even if they're in different formats and places. The aim is to spare doctors and nurses the tedious task of collating the data themselves to get a complete picture of a patient's history.

I Can't Do This All on My Own

The medical profession has had a slightly love-hate relationship with Google over the years. Some (perhaps many) doctors have used Google search to help bolster their diagnoses and to illustrate symptoms patients should look out for. On the other hand, patients coming to doctors with second opinions from Dr. Google can present... challenges.

Now though, generative AI is here to threaten Google's long-held search engine dominance and while Google's Bard has not yet been able to eclipse Chat-GPT, designing bespoke search engines for businesses is an area where Google can potentially keep its edge, and it's going big on healthcare:

  • In April, Google started testing an AI tool for helping to diagnose patients at the Mayo Clinic, and more recently it built an AI to advise humans on whether to trust another AI-powered diagnosis. Hey, you know what might be handy? An AI to check how trustworthy that AI is.
  • Google says the new search tool won't just collate records; healthcare workers can also pose narrower questions such as: "What medications has this patient taken in the last 12 months?" per CNBC.

Internal Cash Bleeding: While AI products abound, they're not necessarily money-spinners, according to a Wall Street Journal report published on Monday. Even major players in the generative AI boom have managed to create money-losers, because although AI chatbots appear to generate magic responses out of thin air, they in fact require a lot of processing power to run. One source told the WSJ that a Microsoft AI product designed to help coders cost the company an average of $10 per user to run, with some users costing it as much as $80 per month. With that kind of burn, Microsoft might need a doctor.