Nine Breakthroughs Made Possible by AI

Article Content

This story originally appeared in the winter 2025 issue of UC San Diego Magazine as “Made Possible By AI, And Those Who Dream Big.”

The rise of artificial intelligence has already transformed every facet of our lives at a remarkable speed. At UC San Diego, AI is empowering bold advances in science, education, public safety, health and the arts.

AI AND HEALTH

From uncovering hidden disease mechanisms to guiding new therapies and medical technologies, UC San Diego researchers are harnessing AI to propel humanity toward a future in which diseases are mitigated or eradicated. What was once science fiction is now possible. As a tool, AI speeds diagnoses, offers a better understanding of the human body and enables more innovative treatments.

AI Helps Uncover Alzheimer’s Trigger

A gene once thought to be just a biomarker for Alzheimer’s disease is actually one of its causes, according to the work of UC San Diego bioengineers. Using AI to model the 3D structures of proteins, the team discovered that the gene PHGDH has a hidden “moonlighting” role: It disrupts how brain cells switch genes on and off, a disturbance that can fuel the disease.

AI Targets Tuberculosis

Tuberculosis remains one of the deadliest infectious diseases, infecting more than 10 million people worldwide in 2022. Drug-resistant strains make treatment increasingly difficult, but improved TB treatments are on the horizon. Joe Pogliano, professor, and Kit Pogliano, dean and distinguished professor, both in the School of Biological Sciences, worked with Linnaeus Bioscience and Seattle Children’s Research Institute to develop MycoBCP, a new AI-powered tool that could accelerate the search for TB treatments. The technology builds on the Poglianos’ earlier bacterial cytological profiling method, which quickly shows how antibiotics work. By pairing that method with deep learning, researchers could detect subtle changes in TB cells that would otherwise escape the human eye — revealing how potential drugs act on the pathogen.

“This is the first time machine learning has been applied this way to bacteria,” says Joe Pogliano. The advance, funded in part by the Gates Foundation, could fast-track the development of urgently needed TB therapies.

AI Peers Into the Heart

By recording signals from outside of heart muscle cells and using AI to reconstruct what’s happening inside these cells, engineers were able to monitor heart activity with remarkable accuracy — without invasive methods or physically penetrating the cells. The breakthrough offers safer, faster insights into how heart cells function, communicate and respond to new drugs.

AI Sharpens Breast Cancer Treatment Plans

Researchers have developed advanced deep-learning techniques that could revolutionize treatment planning for breast cancer radiotherapy — making it faster and improving its quality. The approach reduced errors in radiation doses to critical organs, such as the heart and lungs.

AI in Motion 

UC San Diego engineers have created a flexible patch — a wireless, skin-mounted ultrasound device — that monitors muscle activity in real time to control a robotic arm, among other uses.  To extract additional insights from these muscle signals, the researchers developed an AI algorithm that maps the signals to their corresponding muscle distributions. Compact, battery-powered and designed for long-term wear, the technology could open new possibilities in health care monitoring and human-machine interaction. 

Using AI to Enable Better Vision

In a unique and long-standing collaboration, UC San Diego electrical engineering graduate students have been embedded with the Jacobs Retina Center at Shiley Eye Institute to develop better computer vision, AI and image-processing tools. These will help physicians diagnose patients with greater speed and accuracy, predict the most effective treatments, and aid in the development of new treatments.

“It’s exceptionally rare to have weekly discussions [in which] engineers actively engage with patient care, making it easier to understand their work and needs,” says Dr. William Freeman, a distinguished professor and vice chair of the Viterbi Family Department of Ophthalmology at UC San Diego School of Medicine and director of Jacobs Retina Center at Shiley Eye Institute. “This sustained effort, bridging medicine and engineering across campus, isn’t just about claiming that we use AI — it’s a dedicated approach to tackling  practical health care challenges through innovation.”

AI AND CLIMATE CHANGE

UC San Diego researchers are using AI to better understand Earth’s changing environment. These innovations provide faster forecasts, actionable insights and tools that help protect communities and the planet.

AI Accelerates Climate Forecasts

Running global climate simulations can take weeks on supercomputers, limiting the number of scenarios scientists can explore. Researchers at UC San Diego and the Allen Institute for AI have developed a new model, Spherical DYffusion, that projects 100 years of climate patterns in just 25 hours.

By combining generative AI techniques with physics-based data, the model delivers results 25 times faster than current methods without the need for massive supercomputers. The breakthrough could provide scientists and policymakers with faster and more flexible tools for anticipating the long-term effects of climate change.

AI Helps Fight Wildfires

With a growing network of more than 1,200 natural-hazard monitoring cameras spanning remote mountaintops to wildland-urban interfaces, ALERTCalifornia collects real-time data and utilizes AI that helps emergency managers spot smoke, monitor fires and plan evacuations. UC San Diego’s WIFIRE team at the San Diego Supercomputer Center utilizes those camera and sensor feeds to pinpoint wildfire ignition locations and create predictive models that map how a fire will spread. That valuable information is shared with emergency managers and first responders within minutes via the statewide program known as FIRIS, or the Fire Integrated Real-Time Intelligence System. These models are further refined by AI-powered aerial sensing — including infrared imaging — to track fire perimeters through thick smoke and give responders a strategic advantage as wildfires progress.

Following the LA fires earlier this year, ALERTCalifornia deployed mobile cameras to monitor burn scars for landslides, mudflows and flooding during a forecasted atmospheric river, providing emergency managers with critical, real-time information to inform evacuations.

AI AND ART

“I am a body, and I am also a multitude of smaller, living bodies.

I am a body, and I am also part of a greater living body.”

These are verses from a multiscreen video and sound installation — titled “Superradiance. Embodying Earth.” — from Memo Akten, artist and assistant professor in the Department of Visual Arts, and his collaborator, Katie Peyton Hofstader. It is a peaceful, immersive experience that was made in concert with AI.

Combining dance, poetry, neuroscience and generative AI, “Superradiance” explores embodiment, technology, and planetary awareness and its consciousness. Shape-shifting silhouettes emerge from and dissolve into forests, oceans and deserts, illustrating the interconnectedness of people and the planet. The work was the centerpiece attraction of the exhibit “Embodied Pacific: Oceans Unseen” at Birch Aquarium at Scripps Institution of Oceanography. 

For Akten, AI is not a substitute for creativity but rather a collaborative tool for probing bigger questions: “What is the nature of reality? What is the nature of life? What is the nature of the mind?”

A specialist in new media arts, also referred to as computational art, Akten merges conceptual ideas and his artistic vision with technical experimentation. His process begins with what he calls “high-level conceptual motivations,” or themes that spark his interest, such as the intelligence that emerges from the laws of physics, including how slime mold is able to solve mazes or even how faith and people’s relationships with each other are evolving alongside technology.

He then layers these ideas with computational algorithms — AI — through which he can utilize different techniques, sounds, graphics and media. “I visualize it like stalactites and stalagmites,” says Akten. “On one hand, I have these top-down ideas with no particular idea of how that work might manifest itself. And then, on the other hand, I’m exploring techniques and technical approaches and technologies [to build up to those ideas], and sometimes they connect,” he says. “And when they meet, that becomes an artwork.”

Akten’s innovative approach has earned him the Prix Ars Electronica Golden Nica, and exhibitions at the Venice Biennale, among other accolades.

For Akten, art and AI are not opposing forces but co-creators — each expanding what’s possible.

Learn more about research and education at UC San Diego in: Artificial Intelligence