Researchers use AI to keep kids safe online | Virginia Tech News

Applications of AI


Technology-enabled education

Much of the work to prevent online child abuse focuses on prosecuting perpetrators of cybercrime, which Cho said is helpful but has limitations as children and teens are still at risk of being targeted by abusers.

Cho has been working for several years to develop a project to provide tools and training to help children and teens protect themselves from perpetrators who use the internet to sexually exploit them. It's a challenge, she says, because there are many ethical issues involved in educating kids about abusers and their tactics.

The new project will focus on developing chatbots that simulate children's interactions with online predators. These simulations can be used to develop educational programs, primarily aimed at 11-15 year olds.

Given the challenges, this has been a very collaborative project.

Huang, an assistant professor of computer science, will work on various aspects of the AI ​​that powers the bots.

“I have a very strong interest and strong belief in using AI technology to solve social problems,” Huang said.

He will leverage previous research into conversational AI, which uses large-scale language models to make educational bots as believable as human interactions.

Li, also an assistant professor of computer science, will work on collecting data to train bots, which he said is a particularly difficult problem because it requires finding data about how predators interact online and using it responsibly.

“The challenge with this research is that there is no authentic cybergrooming conversation dataset that can be used to train a chatbot,” Lee said. “We plan to tackle this problem with a human-centered approach and establish an ethical platform where young people and their parents can work together to generate such data and raise awareness against cybergrooming as part of the data collection process.”

Another collaborator is Pamela Wisniewski, an associate professor of computer science at Vanderbilt University and an expert on youth privacy and online safety. With a related $344,874 National Science Foundation grant, Wisniewski will consult with cyber abuse and sex trafficking prevention experts and youth workers to develop training materials to ensure the chatbots are safe for minors to use.

“We hope to design, develop and evaluate tools that can help young people by educating them about the risks of cybergrooming and what they can do to protect themselves from them,” Wisniewski said. “The goal is not to restrict or monitor young people's internet use, but to provide them with the tools they need to be safe online.”





Source link

Leave a Reply

Your email address will not be published. Required fields are marked *