Table of Contents
10 Facts About Stanford’s Human-Centered AI Institute
Stanford University launched its Institute for Human-Centered Artificial Intelligence in 2019. The university has been a leader in artificial intelligence, multidisciplinary collaboration, and diversity of thought, blended together to create the university’s institute.
Directing HAI are both John Etchemendy and Fei-Fei Li. Their vision for the institute is to build a better future for humans by using AI, hence the human-centric side of things. The institute intends to work in various sectors, including finance, tech, healthcare, and manufacturing.
Stanford is well-known for being a high-level university with options for students to study humanities, engineering, medical schools, and social sciences. Many professionals have come from this prestigious school, and it is known for being on the leading edge of new technology. The school has decided to add the Human-Centered AI Institute to continue this.
What Is the Human-Centered AI Institute Focused on for Research?
The Human-Centered AI Institute, or HAI, focuses more on certain research areas than others. They divide their interests into three areas:
Human Impact
AI and humans are frequently mixed, so it’s essential to improve the interactions between the two. HAI works to promote research in this area and ensure AI’s integration into the human world. This means some studies are taking place on many areas of AI use. For example, using artificial intelligence for judicial systems, industry structuring, and population control, etc.
There is a big potential for misusing AI in these areas, among others, so the research conducted by those joining the Human-Centered AI Institute is designed to help create a smoother, problem-free integration of the technology with human work.
Augment Human Capabilities
HAI works to create and develop new tools and design methods for augmenting human capabilities. This is done to improve everything from education and healthcare to sustainability and automation.
Many people are unhappy or unmotivated in their jobs, but AI can create better opportunities to train and educate people in many industries. In addition, it’s possible to improve a worker’s contribution to their job to ensure higher quality work.
The majority of the human impact research done at HAI looks at managing AI’s integration into thehuman workforce. While many people feel that AI can take over human jobs, it’s usually more beneficial to blend the two.
Examples of research projects that have been part of the HAI include a dynamic artificial intelligence therapy for children with autism via Google Glass. The project works to make it possible for those with ASD to understand social emotion cues, reacting more efficiently.
Another project uses haptic feedback for motion guidance, making it simpler to use AI for human-machine interaction. It’s likely to be used for proper communication between humans and machines in cases where robotic surgery is being performed, collaborative object manipulation, and more.
Intelligence
Artificial intelligence is nothing without intelligence, so it makes sense that HAI focuses on creating better machine intelligence. AI currently lacks the smoothness necessary to integrate seamlessly with people.
This area is meant to develop AI that is better at studying, using, and understanding human language, behavior, interactions, and intentions. Ideally, the AI will copy and understand humans to better mimic their reactions. We see attempts at this in search engine algorithms every day, but anyone online knows how those can fail.
AI is all about learning from what it observes. While you need to add some information to help the AI start, it does learn over time. The research programs that focus on the machine-assisted world often look at ensuring the AI makes more human decisions. However, they should also improve upon those while considering human language, emotions, behaviors, etc.
A good example of ongoing research projects in HAI is the Free Exploration in Human-Centered AI Systems. Mohsen Bayati and Ramesh Johari are running this. Their system determines the best way to balance exploration and exploitation to increase and maximize long-term rewards. The idea behind this project is to build a formal methodology for integrating humans and machine learning.
Another option is the adversarial examples for humans. This study takes a closer look at the various aspects of human perception and how they manage things, as opposed to the negative issues faced by the machines. In many cases, AI systems have to overcome many obstacles, trying to recognize the human language, looking at vision, etc. So, where do humans have vulnerabilities? The study will show this process and create an AI system that can work closer to how a human would work.
Fun Facts About HAI
Interested in knowing more about this unique place?
- The institute is run by two co-founders, John Etchemendy, a professor of philosophy and was once a provost at Standard, and Fei-Fei Li. Li is a computer science professor and previously worked as a director of the Stanford AI Lab.
- HAI hopes to become a global hub for everyone interested in working within the AI industry, including researchers, developers, and government officials.
- HAI has a strong focus on ethics and how it applies to artificial intelligence, particularly in healthcare and other areas that may be protected.
The future of HAI is likely to be bright as AI becomes more and more important in the world. Researchers will be needed to find better ways to integrate machine learning with the world of human workforces and create a more efficient workplace.