You can't walk through a healthcare conference these days without being bombarded by Artificial Intelligence. The promises are huge: AI will replace radiologists, cure diseases, and slash costs. After 25 years of listening to these pitches, I've learned that the AI that actually works is rarely the stuff of headlines. It isn't about replacing doctors. It's about giving them better tools to solve the tedious, operational problems that burn them out and get in the way of patient care. Here's what AI looks like when it's actually delivering a return.
The Hype Never Quite Meets Reality
For years, we were told AI would put radiologists out of work (they're now in higher demand than ever). The truth is, the real AI revolution is happening in the background. It's operational. Across our companies, we're using AI to optimize staffing schedules, predict when a machine needs maintenance, or automate the administrative work that drives good clinicians crazy.AI That Actually Pays for Itself
A flashy demo is one thing, but a measurable return is another. Here's where we've seen it work: At FasPsych, our telepsychiatry company, we don't use AI to diagnose patients. We use it to augment their work, helping triage cases and optimize schedules. The result can be a 4-to-1 return on investment and significant savings for hospital partners[7]. The RhythmStar® device from RhythMedix uses AI to analyze heart rhythms, but its real value is that it gives cardiologists clean, actionable alerts instead of a firehose of raw data[8]. Our primary care group, Complete Health, uses a system that doesn't treat a single patient. It analyzes health patterns across their entire patient population to find opportunities for preventive care that keeps people out of the hospital[9].What Consistently Works
Time and again, I've seen the most effective AI fall into three practical buckets:1. Operational AI: The unglamorous work of managing schedules, inventory, and patient flow.
2. Risk-Stratification AI: Finding the small number of patients at the highest risk of a major complication or readmission, so clinical teams can step in before it happens.
3. Administrative AI: The true unsung hero. This is AI that handles prior authorizations and claims, freeing up a nurse to be a nurse, not a clerk.
AI in healthcare isn't about artificial intelligence replacing human intelligence. It's about using it to augment our own so we can provide better care to more people. That's not just good tech—it's a solid investment.References
- FasPsych ROI metrics, https://faspsych.com (Homepage: "For every $1 invested in mental health, employers see a $4 return") ↩
- RhythMedix portfolio description, https://pharosfunds.com/RhythMedix.php ↩
- Complete Health portfolio description, https://pharosfunds.com/complete-health.php ↩