AI that addresses ALS, mental health, refugee flow predictions

BUFFALO, N.Y. – Not all chatbots are created equal.

Some, like those used in customer service, are simple. Others — like the systems University at Buffalo researcher Rohini Srihari builds — perform far more complex tasks, such as giving voice to someone with ALS who has lost the ability to speak.

In the latest episode of Driven to Discover, a University Communications podcast that explores what inspires UB faculty members, Srihari shares how she combines her love of language with computer science to create AI-based tools for the public good.

This includes devices that help people with motor neuron diseases communicate, and tools that address the nation’s mental health crisis.

“People are using chatbots for mental health support, as a conversational partner to overcome isolation and all that. The danger is that these tools have not been vetted for that purpose,” she says. “And that has sort of influenced our work, and our work has shifted from actually creating the chatbots that provide mental health support to coming up with systems that evaluate how good they are.”

Another project she has explored involves using AI to predict the flow of refugees, which can aid governments, nonprofits are other organizations that assist refugees. She’s even working with choreographers to see how AI might help them create new works.

Much of her work is made possible by Empire AI, New York State’s more than $500 million research consortium and supercomputing center, which is located at UB. The computing power of Empire AI, Srihari says, is accelerating important research and bringing society closer to solutions for complex global problems.

“Empire AI has been a tremendously valuable resource for us,” Srihari says. “Something that took us five days, we are now completing in less than a day, in a few hours.”