It’s a great honor to be invited to deliver the Ramon Guiteras Lecture at AUA2024. When I was asked, the theme requested was “what in current technological advances should urologists be concerned about, such as generative AI (artificial intelligence) like ChatGPT?” (If you’ve been living under a rock for the last couple of years and haven’t yet seen the thousands of articles and news features about the arrival of this relatively new form of AI, I encourage you to go to chat.openai.com, try it out for yourself, and see what you think.) Urologists are technophiles, and our approach to new technology is generally sanguine at worst and ardent at best as we incorporate new innovations into our care of patients. We’ve done that with endoscopes, neural stimulators, lasers, microscopes, robots, and much, much more, so my general feeling about ChatGPT and urology is that we’ll find a way to make it benefit urological health. But let’s take a closer look at what ChatGPT is, and even more importantly, what’s new in our innovation toolbox and its education, because there’s a lot going on there.

It’s useful to look back to the birth of modern computation to understand what a large language model like ChatGPT can offer, because what was true almost 200 years ago is highly relevant today. Charles Babbage, a 19th century English mathematician, set out to build a mechanical device to generate tables of polynomials, the “difference engine.” He was unable to complete it due to the primitive craftsmanship of the time. Amazingly, he then set out to design an even more flexible and robust computer that could attack any solvable mathematical problem, the “analytical engine.” It could never be made in his era, but with the advent of the transistor in the second half of the 20th century, analytical engines can now be found everywhere, in our telephones, on our desks, in our cars, and nearly anywhere that benefits from programmable devices. Lord Byron’s daughter, Ada Lovelace, a brilliant mathematician, studied Babbage’s plans for the analytical engine and wrote much about it, in fact writing the very first computer program. But she also opined about its utility and wrote: “It is desirable to guard against the possibility of exaggerated ideas that might arise as to the powers of the Analytical Engine….The Analytical Engine has no pretensions whatever to originate anything. It can do whatever we know how to order it to perform. It can follow analysis; but it has no power of anticipating any analytical relations or truths. Its province is to assist us in making available what we are already acquainted with.”¹

What Lady Ada wrote about the analytical engine is entirely true of ChatGPT. ChatGPT uses large swaths of digitally available material and rearranges them according to the likelihood that words, phrases, and sentences would follow. Although it can sound like us, it doesn’t think like us, and it can’t create in the unique way that biological humans do. So while it can function like a clever human imposter at times, it doesn’t present the full panoply of human intelligence, and we should be able to easily tame this beast of our own construction for our own devices (Figure 1).

Yet what is really promising in modern innovation are the tools that are suddenly ubiquitous, inexpensive, accessible, and easily understood. And we can use them in teaching innovation to our medical students, urological residents, and fellows, providing a powerful future workforce that not only cares for patients, but also creates the devices involved in that care (Figure 2).

One set of tools are small, inexpensive, easily programmed, yet highly powerful computers. The Arduino was invented in 2005 by 2 faculty at the Interaction Design Institute in Ivrea, Italy. It is typically programmed using Processing, an accessible language invented in the MIT Media Lab designed to teach computer programming to non-STEM (science, technology, engineering, and mathematics) students. There is now a family of Arduinos, running upward from $20. The Raspberry Pi was invented in 2006 at the University of Cambridge, and it’s a full computer with USB ports, HDMI for a high resolution monitor, Wi-Fi, and removable disk storage in the form of a micro SD (Secure Digital) card. It sports a full Linux operating system and runs upward from $5. Arduinos and Raspberry Pis have hardware inputs and outputs that can be attached to sensors, motors, displays, and pretty much anything that can be controlled electrically.

Another set of tools are 3D printers, and these are now available to consumers, costing as little as $200. With freely available design software, users can design and print just about any object. Combining the smarts of Arduinos and Raspberry Pis, anyone can create all sorts of machines and devices. (An example is the “Lotion-o-Meter” shown in Figure 3, which I made with a 3D printer in my closet and an Arduino that tells me when I need to apply lotion.)

We use these powerful tools to teach innovation to our urological learners. In a structured curriculum, students go through the process of problem identification, describing it and creating a high-level specification for what is necessary in a solution; intellectual property, market, and existing product research; ideating solutions; prototyping and testing them; and, finally, securing intellectual property for the unique solution. In the modern era, a team of contributors from varying disciplines, medicine, design, engineering, business, and law, work together in education to prepare the learner for a future of solving problems and making these solutions available to all. (An example of our educational group is shown in Figure 2.) It’s an exciting time, and I’d say that the promise of modern innovation in medicine and urology far outpaces the peril.