Healthcare AI today: Trump’s American AI offensive, healthcare’s lack of regulatory cohesion, chatbots’ silence, more

News you ought to know about: 

• President Donald Trump has crystallized his No. 1 priority for artificial intelligence. It’s “winning the AI race.” Those four words are the operative part of the title in a 28-page White House action plan posted July 23 and vocally promoted by the President at an unveiling event that same day. The details of the plan fall under three broad directives, or “pillars.” These are 1.) accelerate AI innovation; 2.) build American AI infrastructure; and 3.) lead in international AI diplomacy and security. Recalling the USA vs. USSR space race of the 1950s and ’60s, the text of the plan refers to the AI “race” 10 times, positioning the Chinese Communist Party as our fiercest rival in the 21st-century high-stakes technology contest. 
   
  • The document leads with a quote from the President giving an overview of the content. “Today, a new frontier of scientific discovery lies before us, defined by transformative technologies such as artificial intelligence,” the quote reads. “Breakthroughs in these fields have the potential to reshape the global balance of power, spark entirely new industries and revolutionize the way we live and work. As our global competitors race to exploit these technologies, it is a national security imperative for the United States to achieve and maintain unquestioned and unchallenged global technological dominance. To secure our future, we must harness the full power of American innovation.”
     
  • To this AI and Crypto Czar David Sacks adds that, to win the AI race, the U.S. “must lead in innovation, infrastructure and global partnerships. At the same time, we must center American workers and avoid Orwellian uses of AI. This Action Plan provides a roadmap for doing that.”
     
  • At the July 23 AI Action Plan unveiling at D.C.’s Mellon Auditorium—co-hosted by the private, pro-business Hill and Valley Forum with the “All-In” podcast, of which Sacks is a co-host—Trump advanced the theme of AI development being a global competition. “Winning the AI race will demand a new spirit of patriotism and national loyalty,” he said. “For too long, much of our tech industry pursued a radical globalism that left millions of Americans feeling distrustful and betrayed. Many of our largest tech companies have reaped the blessings of America freedom while building their factories in China, hiring workers in India and slashing profits in Ireland, all the while dismissing and even censoring their fellow citizens right here at home. Under President Trump, those days are over.”
     
  • To get the action plan rolling without delay, the President signed three new executive orders. Largely aligning with the action plan’s three pillars, the EOs are titled “Promoting the Export of the American AI Technology Stack,” “Accelerating Federal Permitting of Data Center Infrastructure” and—fleshing out Sacks’s distaste for “Orwellian” uses of AI, “Preventing Woke AI in the Federal Government.” 
     
  • Detractors wasted no time calling out the Administration’s big AI day as a sham. A day ahead of the action plan’s release and unveiling, scores of watchdog groups and activist organizations posted a petition for a counter-measure called the “People’s AI Action Plan.” This reads, in part: “We can’t let Big Tech and Big Oil lobbyists write the rules for AI and our economy at the expense of our freedom and equality, workers and families’ well-being—even the air we breathe and the water we drink—all of which are affected by the unrestrained and unaccountable rollout of AI.”
     
  • One of the signatories of the petition statement, the nonprofit consumer advocacy organization Public Citizen, posted its leaders’ reactions to the Trump administration’s AI activities of July 23. The watchdog’s Big Tech accountability advocate, J.B. Branch, doesn’t mince words. “Under this plan, tech giants get sweetheart deals while everyday Americans will see their electricity bills rise to subsidize discounted power for massive AI data centers,” Branch says. “This isn’t leadership—it’s a sellout.”
     
  • Additional coverage is everywhere. 
     
• Three of four stakeholder populations in healthcare AI—providers, payers and consumers—worry about gaps in AI regulation. Only vendors seem unconcerned. Surveyors with Black Book Research uncovered the perhaps-predictable views upon conducting a 360-degree poll. The firm finds 40% of providers viewing current federal oversight as “inadequate to ensure safe and ethical AI deployment in clinical environments.” The vast majority of consumers, 92%, would like to have a national oversight body dedicated to AI in healthcare. In stark contrast, only 7% of vendors support expanded federal regulation. “The healthcare sector is embracing AI at an accelerating pace, but regulatory structures have not evolved proportionately,” remarks Black Book President Doug Brown. “This flash poll underscores the need for multi-stakeholder collaboration to ensure responsible innovation while addressing patient safety, ethical use and trust.”
   
• Medical AI chatbots always used to remind users: ‘Be sure to check my answers to your health questions with your doctor.’ That’s happening less and less, according to a new study out of Stanford. Co-author Roxana Daneshjou, MD, PhD, says the lack of disclaimers may put many people at risk of harm. “There are a lot of headlines claiming AI is better than physicians,” she tells MIT Technology Review. “Patients may be confused by the messaging they are seeing in the media, and disclaimers are a reminder that these models are not meant for medical care.” An MIT researcher not involved with the study suggests AI companies may be skipping the warnings as a way to “elicit more trust in their products as they compete for more users.” 
   
• Most large hospitals in the U.S. now use AI for one or more clinical purposes. Yet few check points of potential legal exposure with a risk manager. They’ll be in a deep hole should an algorithm get implicated in a lawsuit over patient harm. The British-American multinational risk-management firm Willis Towers Watson offers perspective and advice on the matter in a white paper published July 21, “AI in Healthcare: Who pays when AI fails?” The best strategy for managing AI risk in healthcare is to adopt the “tried and tested methodology of assessing, quantifying and mitigating the risk,” the authors write. “While the increased utilization of AI will undoubtedly amplify existing exposures already catered for in established insurance products, in particular cyber risk, it already has and will continue to open up new exposures—some of which may fall within scope of existing insurance policies and some that will not.” Download the paper here. 
   
• Physicians considering AI software for purchase should ask themselves as well as the vendor some pointed questions. What clinical problem does the AI tool solve? What evidence supports the tool’s safety, accuracy and efficacy? How will it integrate into existing clinical workflows? Things like that. Healio spoke with some physician experts on AI for guidance on those three inquiries and three more. By asking the right questions, Deepti Pandita, MD, of UC-Irvine Health says, “physicians maintain their role as stewards of patients.” Read the piece.
   
• Also worthwhile: 
   
  • Where are all the AI drugs? (Wired)
     
  • The world’s fastest-growing healthcare AI market: Middle East/North Africa (WAYA Media) 
     
  • AI jumps ahead of EHR in health system tech priorities (Modern Healthcare, behind paywall)
     
• From AIin.Healthcare’s sibling news outlets:
   
  • Health Imaging: GPT-4o’s ‘all or nothing’ accuracy continues to hinder its radiologic capabilities
     
  • Cardiovascular Business: New data highlight how Heartflow’s AI software drives major shifts in heart care
     
  • Radiology Business: Imaging AI vendor files for $100M IPO
     
  • Health Imaging: AI ups detection of small prostate lesions by nearly 20%