Remember a few years back when the AI boom hit? Everyone was talking about how robots were going to take over the world and replace all our jobs, including our doctors.
Fast forward to 2026. Did a robot scan your throat at your last check-up? Nope. But behind the scenes, after a solid few years of real-world testing, AI has quietly pulled off some real but uneven medical progress, mostly as a support tool rather than a revolution in everyday clinical practice.
Let’s look at where the world is right now with this technology, and how Croatia is actually keeping up.
The Global Stage
While we were busy worrying about chatbots taking over the internet, medical researchers were using AI to do things that sound completely like science fiction. One of the most incredible things AI is doing today is helping decode brain signals for communication. For patients who are completely paralyzed and unable to speak due to severe strokes or Amyotrophic Lateral Sclerosis (ALS), the same condition physicist Stephen Hawking had, scientists have implanted tiny sensors in their brains. In the past, Hawking had to rely on a tedious system where he twitched a cheek muscle just to select words on a slow-moving screen, managing only about 1 to 15 words per minute. Today’s AI in experimental systems can bypass physical movement in controlled laboratory settings and translate brain activity into text or synthetic speech. In some research demonstrations, performance can approach tens of words per minute and in some lab setups even higher, allowing patients to communicate with their families at speeds of up to nearly 80 words per minute in some experimental systems, just by thinking about the words, although this is still not widely available in everyday clinical use.
Then there is the fight against superbugs, those terrifying bacteria that do not respond to any known antibiotics. Discovering a single new antibiotic traditionally takes over a decade of trial and error. Today, researchers at top institutions like MIT and others are using generative AI to digitally design and screen tens of millions of chemical compounds. In some cases, AI systems have helped identify promising antibiotic candidates in days or weeks instead of years. However, these are early-stage discoveries that still require long laboratory and clinical testing before becoming real medicines. The AI acts like a super-fast matchmaker, narrowing down huge chemical spaces to find potential drug candidates much faster than traditional methods.
Even cancer treatment has been significantly upgraded in research settings. Instead of giving a patient standard chemotherapy and hoping for the best, AI models can analyze tumor data including genetic and imaging information to help predict treatment response. Some models show promising accuracy in predicting which therapies are more likely to work, but in real-world clinical practice this is still an emerging tool rather than a fully standardized clinical decision system. The goal of this approach is precision medicine, where treatment is increasingly tailored to the individual patient rather than a one-size-fits-all approach.
Meanwhile in Croatia
Okay, so the world is doing some heavy sci-fi stuff. Where does Croatia stand in all of this?
We might not be building brain-reading microchips just yet, but Croatian hospitals have quietly been introducing modern AI-assisted imaging tools as part of broader digital upgrades in healthcare. The focus here has been on practical, everyday situations where saving time is critical for the patient.
One of the best examples is cancer detection. Clinics in Zagreb now use advanced CT and PET/CT systems with AI-based software that can help analyze scans and highlight suspicious tissue changes that may be difficult to notice. These systems can assist in detecting small lesions, helping radiologists focus on areas that deserve closer attention. Instead of simply producing images, the scanners can support specialists by acting as a second set of eyes, although final interpretation always remains with the doctor.
AI has essentially become a second pair of eyes for radiologists. A modern CT scan can generate hundreds or even thousands of images. Reviewing every single image manually takes time and concentration. AI software can rapidly screen those images, flag unusual patterns, and help doctors prioritize the most urgent findings. The technology does not replace specialists, but it helps them work faster and reduces the chance that subtle abnormalities go unnoticed.
The same idea is now appearing in preventive medicine. During routine colonoscopies, some clinics worldwide and in parts of Europe, including pilot or early adoption settings, use AI systems that analyze the live video feed from the endoscope in real time. As the doctor moves through the colon, the software can help detect and highlight tiny polyps and suspicious tissue growths. Small polyps that might once have been difficult to detect can now be identified and removed before they have the opportunity to develop into colorectal cancer.
This may not look as futuristic as reading brain signals or designing drugs with artificial intelligence. But for a patient whose cancer is discovered months earlier, or whose precancerous polyp is removed before it becomes dangerous, the impact can be just as life-changing.
Conclusion
Ultimately, the prediction that AI would replace doctors has so far proven wrong. Instead, it has become one of the most powerful tools doctors have ever had. From helping paralyzed patients communicate in experimental systems, to accelerating early stage antibiotic discovery, to assisting in medical imaging analysis in hospitals, AI is not replacing medical professionals. It is amplifying their abilities.
The future of medicine is not humans versus machines. It is humans and machines working together to deliver faster insights, better supported diagnoses, and improved outcomes for patients.
