Michigan Engineering News

A graphic design featuring a series of circles connected by lines, representing a neural network.

Artificial Intelligence

Michigan Engineers are at the forefront of advancing artificial intelligence, from its core enabling technologies to impactful, interdisciplinary applications that address today’s complex challenges. Our pioneering research and education builds on longstanding leadership that goes back to the dawn of the information age.

With the Michigan AI Lab as the foundation, our faculty continue to advance the models that underpin all AI technology, including generative AI, machine-learning and deep-learning models that uncover patterns in complex data, as well as the computer-vision and natural language models that serve as the “brains” of robots that can learn from people and adapt to new situations.

Our faculty partner across the University, harnessing U-M’s breadth of expertise to explore new applications and integrate this technology into critical areas like healthcare, through the e-Health and AI partnership with Michigan’s prestigious hospital, as well as energy, mobility, semiconductors and more. They also work closely with industry to improve AI performance and implementation, and they advise policy makers on how to ensure AI models are accurate, trustworthy and sustainable. We educate graduates that go on to be world leaders in AI research.

An abstract digital circuit board with vibrant neon colors, featuring a complex grid and interconnected shapes.
A computer rendering of computer internals, lit up by neon

The planned facility for high-performance computing and AI research has secured $100M from the state.

Explore more AI news

An abstract digital circuit board with vibrant neon colors, featuring a complex grid and interconnected shapes.
Portrait of Karthik Duraisamy.

The director of the Michigan Institute for Computational Discovery and Engineering discusses the institute’s past and future.

Explore more AI news

Faculty at Michigan Engineering have been at the forefront of machine learning and artificial intelligence research, and our graduates and professors started several research fields that served as the foundation for modern artificial intelligence. Claude Shannon, who graduated from Michigan Engineering in 1936, started the field of information theory and became widely known as the “father of the information age.” Professor John Holland developed genetic algorithms and learning classifier tools—two tools that are now the current standard for machine learning and computational science. His 1975 book, titled “Adaptation in Natural and Artificial Systems,” remains one of the most frequently cited works on artificial intelligence. Computer chips have the power required to train today’s massive AI models, thanks to the very large-scale integration technology developed by Professor Lynn Conway around the same time as Holland’s book.

We continued to advance the field with the founding of the Michigan AI Lab in 1988. The lab started as a collection of faculty interested in computer vision, but by the ’90s, the lab had attracted a critical mass of researchers working on several major problems in AI. It was one of the larger AI labs at the time, and it explored some of the broadest research avenues.

Several active researchers in the AI Lab and throughout Michigan Engineering are currently among the top one percent of the world’s most cited computer scientists. Michigan Engineering has also trained students that are now world leaders in AI research today, including Waymo co-CEO Dmitri Dolgov, Jordi Ribas, the Corporate Vice President of Search & AI at Microsoft, Peter Wurman, who co-founded the company that became Amazon Robotics, and Kunle Olukotun, a Stanford professor and founder of SambaNova—a company developing some of the fastest AI chips and inference servers.

Today, Michigan Engineering continues to advance the capabilities of AI for both public use and scientific discovery. They work closely with industry experts to improve AI model designs, improve how AI is implemented, and explore new application areas. They also advise policy makers on how to ensure AI models are accurate, trustworthy and sustainable. More than 120 Michigan Engineering researchers are affiliated with the Michigan Institute for Computational Discovery and Engineering (MICDE), where they are working toward a future where AI can help answer some of the most pressing scientific questions of the time. They are also partnering with Los Alamos National Laboratory on a $1.25 billion initiative to build a state-of-the-art supercomputing facility in Ann Arbor.

A computer screen showing pathing options for an AI home wheelchair

Michigan AI Lab

The AI lab brings together award-winning experts to develop new thinking and applications for AI.

A hand gesturing over a computer model on the screen

Michigan Institute for Computational Discovery and Engineering (MICDE)

MICDE supports research that uses generative AI models to drive scientific discovery from large amounts of unstructured data.

A computer rendering of computer internals, lit up by neon

Los Alamos National Laboratory collaboration

In partnership with LANL, we will build a supercomputing facility that will develop AI models for defense and other research applications.

A computer graphic rendering of a neural network

Michigan Institute for Data & AI in Society (MIDAS)

MIDAS enables transformative use of AI across a wide range of research domains.

A roundabout with a large blue circle and yellow, University of Michigan block "M" in the center, surrounded by roads and greenery.

Mcity

MCity and the U-M Ford Center for Autonomous Vehicles provide resources to improve the AI models behind autonomous vehicles.

Eric Krotkov, advisor for the University Research Program at Toyota Research Institute (TRI), center, is given a project update in the lab of professors Dmitry Berenson and Nima Fazeli in the Ford Motor Company Robotics Building.

AI Connections Program

The AI Connections Program facilitates and enables formal collaborations between academic researchers and industry partners.

Surgeons work wearing glasses developed by Michigan Engineers

e-Health and Artificial Intelligence (e-HAIL)

e-HAIL supports collaboration, grant development, and infrastructure for multi-disciplinary approaches to using AI and machine learning in healthcare.

A computer rendering simulating a neural network

Eric and Wendy Schmidt AI in Science Postdoctoral Fellowship

U-M is one of nine universities with funding that specifically supports postdoctoral researchers who are developing AI models for scientific discoveries.

Computer Science and Engineering

  • Improving the capability, accuracy, transparency, and sustainability of generative-AI and computer-vision models
  • Designing AI-powered accessibility tools for people who are deaf and blind
  • Designing AI that powers autonomous vehicles, predicts pandemics and illness, and models business trading agents

Aerospace Engineering

  • Powering and operating autonomous drones
  • Adapting to unanticipated changes and emergency flight conditions
  • Designing new molecules for batteries
  • Virtually experimenting with space propulsion systems
  • Enabling autonomous satellites to dock with other satellites

Chemical Engineering

  • Deciphering how the immune system behaves at the single-cell level to personalize medical treatments and predict therapeutic effects
  • Enabling quantum-mechanical and molecular simulations of materials used for producing chemicals, water treatment, fuels, fertilizers, and medicines
  • Designing new nanoparticle shapes and molecular structures for medical and communication technology

Biomedical Engineering

  • Discovering new medicines
  • Advancing treatments and neuroscience
  • Simulating microbes that are difficult or dangerous to grow and research
  • Allowing patients to control prosthetic limbs with their thoughts

Electrical and Computer Engineering

  • Designing hardware for more powerful and energy-efficient AI
  • Developing new generative AI models, wearable AI, and coachbots that help students learn computing
  • Developing machine-learning tools that create 3D images from x-rays, medical scans, and sound waves

Robotics

  • Building the brains of autonomous vehicles and robots
  • Allowing robots to “see” by processing and interpreting camera and sensor data
  • Enabling robots to learn from human and video demonstrations and adapt to new situations