I was recently in Phoenix for a conference, and while there, I had the chance to learn more about Waymo, the self-driving car company. Standing at the curb, I watched a few Waymos glide silently down the street. My colleagues and I debated: should we download the app and try it, just for the experience? In the end, we chose the faster, familiar option: an old-fashioned Uber.

But as the Waymo disappeared into traffic, I was left wondering: what does this moment say about intelligent decision-making in healthcare?

Waymo’s Intelligent Decision-Making

Think of how a Waymo works. Its sensors capture massive streams of data, lidar mapping, cameras detecting pedestrians, algorithms predicting the movement of every cyclist and car. The vehicle doesn’t just “see”; it interprets, anticipates, and acts.

And the human seated in the back (or the passenger seat) is able to follow right along using the map provided as the Waymo “driver” navigates streets, pedestrians, detours and even unexpected situations (such as accidents on the road). 

Waymo, a subsidiary of Alphabet (Google’s parent company), has been developing and deploying autonomous vehicles for over a decade. Today, its cars have logged more than 100 million miles on public roads and billions of simulated miles, refining the AI that guides their every turn and stop. This AI must make split-second decisions to safely navigate complex urban environments, interpret unpredictable human behaviors, and prioritize safety under uncertainty. Despite our collective nervousness to try Waymo that day in Phoenix, the interesting point is that Waymo reports an 88% reduction in serious injury crashes compared to human-driven cars on the same roads.

What sets Waymo apart goes beyond sensors and software. It’s the intelligent ecosystem powering those cars: the continuous integration of real-world data, real-time decision-making, and rigorous safety validation. The AI isn’t a static program but a learning system that adapts to new environments and collaborates safely with human drivers and pedestrians.

Drawing Parallels: Lessons for Intelligent Decision Making in Healthcare

Healthcare, like urban driving, is a high-stakes environment where decisions must be made quickly and accurately amid complexity. Clinicians juggle massive volumes of patient data, shifting conditions, and ethical considerations daily. Here too, AI promises to support intelligent decision making—not by taking over care, but by freeing clinicians to focus on nuanced human judgment where it matters most.

Waymo’s cutting-edge AI infrastructure offers a metaphor for building the future of digital health technology: systems that work collaboratively, adaptively, and safely within complex real-world environments. Here are a few lessons for digital health:

🧠 Adaptive Edge Decision-Making: Waymo cars rely on “edge” computing: processing data locally on the car to make immediate, context-aware decisions. Similarly, digital health devices and platforms can offer real-time patient monitoring and interventions tailored to the individual, right where care happens. 

🧠 Learning From Experience: Both Waymo’s cars and AI health algorithms rely on extensive training data, whether that’s driving miles or clinical records, to continually improve performance. 

🧠 Continuous Learning Loops: Autonomous vehicles constantly collect data for improving models and safety protocols. Healthcare AI must embrace similar feedback loops, learning from diverse patient populations and clinical outcomes to reduce biases and errors. 

🧠 Ethics and Safety by Design: Just as Waymo publishes extensive safety data and adheres to strict ethical guidelines, AI in healthcare must be developed transparently with robust safeguards to protect patient privacy and autonomy. 

🧠 Collaborative, Human-Centered AI: The best AI systems complement rather than compete with human expertise, providing clinicians with actionable insights while allowing for compassionate care. 

What This Means for Patients, Caregivers, and Clinicians

Many patients and clinicians may feel cautious about embracing AI, much like I hesitated to ride in a Waymo on that Phoenix street. But the story of autonomous vehicles teaches us that intelligent decision making powered by AI can reduce risks, improve safety, and enhance consistency in complex, high-stakes scenarios.
Intelligent decision-making doesn’t mean replacing human judgment. It means integrating data-driven insights into the human decision loop. In fact, the best systems are “human-in-the-loop” models, where AI augments rather than overrides clinicians. Research shows that AI can outperform or match human experts in areas like dermatology, radiology, and pathology when used correctly. But the real power comes when AI and humans work together, leveraging complementary strengths.

For patients and caregivers, this might mean trusting a digital health app to flag early warning signs, such as arrhythmia. For clinicians, it may involve letting AI prioritize imaging scans or suggest treatment options. For health systems, it means designing workflows that don’t just drop a “black box” algorithm into practice, but instead foster transparency, explainability, and shared decision-making.

But there’s a catch: just as my group chose Uber over Waymo for the sake of familiarity and speed, clinicians and patients often default to traditional choices, even when AI-enhanced options could yield better outcomes.

The Road Ahead: Building Trust

Self-driving cars and digital health share a common challenge: trust. Even when the technology is available, reliable, and sometimes even safer than human-driven alternatives, we hesitate. It’s not just about whether the AI works. It’s about whether we, as humans, are ready to rely on it when the stakes are high.

Healthcare today is experiencing its own “Waymo moment.” Artificial intelligence (AI) is increasingly embedded in clinical care: reading imaging scans, predicting disease risks, and even drafting clinical notes. Digital health tools are collecting continuous data from wearable sensors, remote monitors, and apps. Yet adoption lags, not only because of technical limitations, but because patients, caregivers, and clinicians face a deep human question: when do we trust the machine to drive the decision.

Trust in AI (whether in cars or clinics) is built gradually. Studies show that acceptance grows when systems demonstrate safety, reliability, and explainability. Importantly, intelligent decision-making requires balancing efficiency (the “fast Uber”) with innovation (the “new Waymo”). Sometimes, the fastest choice isn’t the most forward-looking one.

Our “Waymo Moment” in Healthcare

That day in Phoenix, choosing Uber over Waymo seemed practical (none of us wanted to download yet another app since we didn’t have Waymo in the different cities we lived in). But in medicine, defaulting to the “old way” may cost opportunities for earlier diagnoses, safer treatments, or more personalized care. The challenge for patients, caregivers, and clinicians alike is to decide: when do we let AI take the wheel, and when do we keep our hands firmly on it?

The answer isn’t “always” or “never.” Like most things in life, it is more nuanced. It’s about learning to co-pilot—using intelligent decision-making to blend the best of human intuition and empathy with machine precision.

Photo by Aamy Dugiere on Unsplash