It’s easy to forget that generative AI launched into the economy—and the public imagination—less than two years ago. How could it be otherwise? Much has happened since OpenAI got the ball rolling with ChatGPT in late 2022.

Healthcare has done its best to surf the aftershocks. To gauge the sector’s degree of adaptation, McKinsey took healthcare’s pulse at two points in time—the fourth quarter of 2023 for baseline and the first quarter of 2024 for trend detection.

Both exercises yielded illustrative snapshots rather than comprehensive overviews, as the firm surveyed 100 representative U.S. healthcare leaders for each.

In a report on the project released last week, McKinsey analysts lay out five sets of results and observations. Here are excerpts.

1. Most healthcare organizations are at least pursuing gen AI proofs of concept.

In Q4 2023, 25% of respondents said they had already implemented gen AI. The count grew to 29% as of Q1 2024.

Despite U.S. healthcare’s general interest in using AI, a substantial portion of respondents is still operating without any plans to pursue gen AI or still maintaining a wait-and-see approach.

2. Healthcare orgs that are already implementing gen AI do so primarily through cobuilding partnerships.

McKinsey’s Q1 2023 survey found 59 of 100 orgs partnering with third-party vendors to develop customized solutions. That number dropped to 42 by Q1 2024, but the count of orgs procuring gen AI products that require limited customization swelled from 17 to 41.

Among those who haven’t yet implemented gen AI, 41% say they intend to buy gen AI products. This behavior may be driven by this population’s concerns with risk (57% are not pursuing gen AI because of risk considerations) and technology needs (29%).

3. Among early gen AI implementers, few have quantified the technology’s impact.

However, 58% believe it is producing a positive ROI.

As with any investment, it’s critical for stakeholders to be able to realize the value that gen AI promises. A measurable positive impact serves as strong reinforcement for continued and expanded use and investment.

4. Surveyed healthcare leaders believe gen AI’s greatest value will come on two fronts.

By name, the two are boosted clinical productivity and patient engagement.

Expectations are also high around gen AI’s potential to improve administrative efficiency and care quality.

5. The No. 1 challenge for healthcare organizations pursuing gen AI is risk.

Not far behind are insufficient capability, data and tech infrastructure, and proof of value.

This demonstrates healthcare organizations’ limited tech readiness to deploy gen AI solutions and also to validate its capabilities.

The report’s authors comment that, as gen AI deployment progresses, healthcare organizations will likely focus on using the technology to support “clinically adjacent” applications.

“However, as organizations develop strong competencies in governance and risk management,” they add, “we expect additional focus on core clinical applications as well, further improving the overall patient/member experience.”

Read the whole thing.

How's ambient AI transforming the practice of medicine? - Clinicians love using ambient AI, but don't take our word for it - take theirs. Head over to Nabla's Wall of Love to see first hand why clinicians love using AI to bolster their day-to-day practice.

Buzzworthy developments of the past few days.

• Here’s where quantum computing and AI meet biomedical research and patient care. It’s a digital intersection. The avenues connect Cleveland Clinic in Ohio with the Novo Nordisk Foundation in Denmark. Together the two institutions are offering a fellowship program that will support 12 researchers over three years with $6.2 million. Recipients get to choose which site they want to work from. Novo Nordisk senior VP Lene Oddershede says the hope is that the tech-heavy work will “contribute to the understanding of disease and improving patient outcomes.” Learn more.
   
• Contrary to contemporary cockiness, healthcare AI is just centuries-old statistical methodology parading around in a robot suit. OK, that’s a bit of an oversimplification. But only a bit. Hear it from a biostatistician and health data scientist. “When trying to incorporate emerging new technologies like generative AI into your organizational processes,” writes Justin Manjourides, PhD, of Northeastern, “it is important to not lose sight of all we have learned about the scientific process of generating reliable results.” With that as a starting point, the good professor offers five points to ponder from statistical science when using AI to inform healthcare research. Healthcare IT Today published the piece Aug. 5.
   
• Want to know another old thing made new by AI? It’s the EEG test. That’s the one in which they affix electrodes on the scalp to monitor brain activity. It’s been around for something like 100 years. And now it’s being combined with AI to help neurologists spot early signs of dementia. The EEG-AI combo may be a quicker first exam than brain imaging, lab testing or cognitive batteries. It’s under development at Mayo Clinic. Read about it here.  
   
• Meet 100 of the most influential AI thinkers in the country. At least four work for healthcare provider organizations, like Divya Pathak, chief data officer for NYC Health + Hospitals. A handful more are with healthcare suppliers, like Parminder Bhatia, chief AI officer at GE HealthCare. The list was compiled by Aim Research and is posted here.
   
• In the U.K., one person of every six thinks healthcare AI will worsen care quality. Surveyors there also found big gaps related to sex and age. Women are a lot less likely to think AI will improve care quality (28%) compared with men (39%). And teens and young adults (ages 16 to 24) are less trusting in the technology’s promise than all other age groups. There’s still plenty of support to go around, though. Check out the full findings.
   
• The U.S. has its share of healthcare AI skeptics too. One of them is a patient who suffers from chronic pain, takes controlled-substance meds to get through it and has to go through predicable rigamarole with every refill. The patient expects the trouble will only get worse now. “AI can’t solve problems that exist on purpose,” the anonymous patient writes in a piece published by the Pain News Network. “Unless patients are at the forefront of the conversations about these tools, I fear they’ll only be used to solve the sole problem hospital administrators actually care about: how to make more money.”
   
• Elon Musk is turning up the heat on OpenAI. This week he filed a lawsuit accusing Sam Altman and Greg Brockman of deceiving him into helping them create the company. The development is noteworthy not only for its vitriol but also for some of the words Musk and his lawyers are using to frame the complaint. The OpenAI guys “preyed” on Musk’s humanitarian concerns, the suit reads before adding: “The perfidy and deceit are of Shakespearean proportions.” Meanwhile Musk lawyer Marc Toberoff tells Fortune: “Even to this day defendants cloak themselves in altruism—but as the world now knows the Emperor has no clothes.”
   
• An AI-equipped robot has performed a fully automated dental procedure on a human. And as if that’s not exciting enough, Mark Zuckerberg’s dad, a dentist, is a key investor in Perceptive, the company behind the technology. “The robotics system has been designed and rigorously tested to ensure that dentists can perform treatments safely, even in conditions where patient movement is prevalent,” Edward Zuckerberg, DDS, assures the press and public. The technology is not yet commercially available. Why am I relieved to hear that?
   
• Recent research in the news:
   
  • University of Colorado: AI for mental health screening may carry biases based on gender, race
     
  • German Cancer Research Center: Prostate cancer: Can AI help to avoid unnecessary biopsies?
     
  • Osaka Metropolitan University: Healthy AI: Sustainable artificial intelligence for healthcare
     
• AI funding news of note:
   
  • Spring Health announces $100M series E funding to accelerate growth and expand global access to mental healthcare
     
  • Billion-dollar investments drive AI surge in healthcare (Pymnts)
     
  • Google’s deal for Character.AI is about fundraising fatigue (Axios)