In 1796, Edward Jenner administered the first ever successful vaccine. Smallpox, one of the deadliest diseases ever, caused an estimated 300-500 million deaths in the 19th century. It is now the only human disease to have been completely eradicated through vaccination. Today, AI is being used to create drugs and, like the vaccine, cure diseases, marking a major advancement in medicine and healthcare. 

The First AI-Developed Drug to Reach Human Trials

Isomorphic Labs, a subsidiary of Alphabet, recently revealed that it was preparing to begin the first human clinical trials of medicines designed entirely by artificial intelligence. Colin Murdoch, the President of Isomorphic Labs and Chief Business Officer at Google DeepMind, told Fortune that the company is gearing up to test its AI-created drugs on real, human patients. These drugs for cancer were developed through human-AI collaboration. 

The company, which spun out of Google DeepMind in 2021, has already established major partnerships with pharmaceutical giants Novartis and Eli Lilly. In March 2025, Isomorphic Labs raised an impressive $600 million in its first external funding round, signaling strong support for its AI-driven approach to medicine. 

The AI models used by Isomorphic Labs can analyze vast amounts of biological data, identify drug targets, and design new compounds. Their model is a more refined version of AlphaFold, an AI developed by John Jumper and 2024 Chemistry Nobel Prize Winner Demis Hassabis. AlphaFold can predict individual protein structures and model complex molecular interactions. Inspired by this, Isomorphic Labs enhanced its functionality, allowing it to model interactions between proteins and molecules like DNA and drugs. Its evolution was essential to its role in streamlining the drug discovery process and accelerating the development of new medicines. 

Isomorphic Labs is placing all its bets on AI, which it believes is the key to unlocking drug discovery at a rate and scale unlike anything seen before. Its mission is to “solve all diseases” with AI, and human trials for its AI-developed drug are just the first step. 

AI in Medicine

Aside from drug development, AI has shown signs of progress in other areas of medicine, including diagnosis, screening, and treatment. Microsoft’s new AI Diagnostic Orchestrator (MAI-DxO) is a generative artificial intelligence tool used to accurately support complex medical diagnoses. Acting as a “virtual panel of diverse physicians,” the system takes on the role of multiple general and specialized doctors -- dubbed in a new report as “medical superintelligence”. The Microsoft report claims that their AI tool is around four times more accurate than a human physician in diagnosing complex issues--and it does so at a lower cost.

MAI-DxO was paired with OpenAI’s 03 reasoning model to understand complex topics. The pairing creates an AI-generated panel of doctors to ask questions, order medical testing, and provide a diagnosis based on follow-ups. When compared with the diagnostic results of 21 human physicians in 304 cases, MAI-DxO solved 85.5% correctly, while the real-life physicians only solved 20% correctly. The disparity can be largely attributed to the physician’s lack of access to books and tools they’d typically use, but this highlights how the AI can speed up the healthcare process and decrease costs for patients. 

Furthermore, the report reveals that Microsoft’s AI consumer products see more than 50 million health-related queries every day, emphasizing the growth of the digital health market. 

Another AI device called iSeg, which can perform tumor segmentation, was developed by Northwestern Medicine. Tumor segmentation, which involves identifying and outlining tumors within medical images, is a complex and challenging process for doctors. It can take multiple visits and scans, and lots of time and energy for a patient.

iSeg employs 3D imagery to clearly understand the tumor -- even as the patient moves or breathes. The AI mapping can also expose areas that doctors may miss with manual segmentation. These areas are usually critical to diagnosis and determining the outcome for a patient. AI technology not only helps doctors treat cancer even more precisely but also closes socioeconomic gaps in healthcare by providing easier diagnoses to certain underdiagnosed groups. 

AI has made strides in medicinal research, development, and diagnosis already, but further testing is needed, and limitations still exist. AI developments have yet to pass clinical trials. Once they do, they can be used in real medical situations, where consistency and accuracy are not just a wish but a necessity. 

Implications and Future Use

As the medical industry approaches an inflection point, AI is redefining the boundaries of care. While models like those from Isomorphic Labs and Microsoft continue designing drugs and making diagnoses more accurately than doctors, medical innovation is shifting to algorithms and technology. All eyes are on the future, where medicine may be led by data-driven AI rather than experience-driven doctors. 

At the same time, AI is collapsing the timeline of medical progress. What once took decades is now feasible in months. At this rate, AI will redefine what it means for a disease to be “incurable” or an issue “undiagnosable.” The future of healthcare is in AI, and the future of AI is in our hands.