Industry Watcher’s Digest
Buzzworthy developments of the past few days.
- The pandemic made freelancers of a lot of formerly fulltime nurses. Since then, many have reported enjoying if not out-and-out preferring their role as gig workers. But now the bloom may be coming off that rose—and it’s AI chatbots doing the wrecking. Researchers at the Roosevelt Institute found this out when they interviewed 30 or so contract nurses and learned the bots have a suspiciously consistent penchant for encouraging independent nurses to “work for less pay,” failing to “provide certainty about scheduling and the amount or nature of work,” and taking “little to no accountability for worker safety.” What’s more, the sample group revealed, the bots sometimes seem to behave like saboteurs, “placing nurses in unfamiliar clinical environments with no onboarding or facility training.” Roosevelt released its report Dec. 17. Read the whole thing.
- Overall, however, nurses are making their peace with AI. Almost half surveyed by the healthcare consultancy Jarrard feel AI is generally a good thing and will either “help nurses like me be more effective in our jobs” or “help address the nursing shortage by handling some tasks nurses currently do.” About 20% disagree, seeing AI as a generally bad thing—one that stands to cull the nurse workforce. The largest single portion of all surveyed nurses, 34%, are undecided about AI’s positive vs. negative ratio. Jarrard analysts also found nurses have a favorable view of workplace technologies in general, appreciating in particular those that improve communications. Survey report here.
- HHS’s 2024 AI Use Case Inventory is up and ready for review. The resource has expanded by 66% and now presents some 271 use cases. That’s up from 163 last year. A blog post introducing the update notes that its use cases are of various sizes and maturities, and that many include plenty of info beyond just summaries. The post is from Steven Posnack, one of whose HHS titles is principal deputy national coordinator for health IT. Full inventory here.
- UnitedHealth is back in the news. And again it’s not for a happy reason. With the company still reeling over the Dec. 4 murder of CEO Brian Thompson, its Optum division has had to clamp down on an internal AI chatbot. That’s because a cybersecurity team found the bot was open for the tampering by anyone with a web browser. The conversational chatbot’s intended purpose is to guide Optum employees in various aspects of claims processing. TechCrunch notes that the tool does not appear to have contained or produced sensitive personal or protected health information. Still, its clumsy exposure “comes at a time when [Optum’s] parent company faces scrutiny for its use of artificial intelligence tools and algorithms to allegedly override doctors’ medical decisions and deny patient claims.”
- Watch for GenAI to seriously up its game across healthcare in 2025. That tip could have come from even a casual observer, no doubt, but it’s worth noting because its source is one of the leading lights in healthcare AI. Aashima Gupta, global director of healthcare solutions at Google Cloud, tells Forbes the rate at which healthcare organizations are investing in the technology looks an awful lot like the frenzied race to get online in the early days of the internet. She backs the observation with five “core trends” that she believes will prove “pivotal at the intersection of healthcare and AI” in 2025.
- Or maybe the mad dash will stall out. Not only in healthcare but across all industries. The Gallup organization finds this could already be happening. Close to 7 in 10 workers tell the pollsters they never use AI, and only 1 in 10 says they use it at least weekly. What’s more, those figures have now held steady for two consecutive years. “This could be a sign that leaders’ aspirations and vision for using AI in the workplace have not yet translated to clear direction or support for employee adoption,” Gallup analysts write in a Dec. 16 report. “Some employees will be early adopters, but many won’t feel comfortable using AI at work until they receive a clear plan and training.” Read the whole thing.
- Here’s something that might be a bellwether of things to come. Or a warning of flops to avoid. Time magazine has introduced an AI chatbot to take readers’ questions about its person of the year (who this year happens to be President-Elect Donald Trump). The bot was trained on the magazine’s own content, along with “other trusted sources” and the bot’s “built-in general knowledge.” Time says the AI has been engineered to “prevent users from steering the conversation toward unrelated, controversial or potentially biased topics.” We’ll be eager to see how that goes.
- Readers of a certain age may be tempted to sing this phone number to the tune of the chorus in the big ’80s hit ‘Jenny Jenny.’ All they’ll need to do is replace 867-5309 with 1-800-ChatGPT. See? The syllable count is perfect. And now try to get that earworm out of your head. Or call the number. Bot operators are standing by. No, really.
- Recent research in the news:
- Stanford: Scientists call for all-out, global effort to create an AI virtual cell
- Northwestern: AI tool analyzes placentas at birth for faster detection of neonatal, maternal problems
- University of Edinburgh: AI-powered blood test spots earliest breast cancer signs
- King’s College London: AI-based ‘aging clocks’ use blood markers to predict health and lifespan
- Stanford: Scientists call for all-out, global effort to create an AI virtual cell
- Funding news of note:
- From AIin.Healthcare’s news partners:
Dave P. has worked in journalism, marketing and public relations for more than 30 years, frequently concentrating on hospitals, healthcare technology and Catholic communications. He has also specialized in fundraising communications, ghostwriting for CEOs of local, national and global charities, nonprofits and foundations.