Can Scientists Build a Lie Detector for AI Chatbots?
Northeastern University researchers are working to improve AI interpretability to better understand how chatbots generate responses. The project aims to demystify the 'black box' nature of neural networks to identify when and why AI models produce false information.
Why it matters
As AI integration grows, the ability to verify the accuracy and logic of automated systems is critical for safety and accountability.
Northeastern professor David Bau is studying how to make AI more interpretable and developing tools to make it easier to tell when it lies.
One of the biggest issues with AI chatbots today is that they can still share false information, according to Northeastern experts.
And researchers still don’t really understand how to get to the bottom of a chatbot’s miscalculation or potential deception, because they don’t have a strong grasp on how AI models produce their answers, the researchers say.
Despite advancements made in computing power, higher quality training data and new reinforcement learning techniques, AI models largely remain an enigma.
“When we use these systems, we don’t know what their algorithms are,” said David Bau, a professor in the Khoury College of Computer Sciences.
Get smarter about the news
Sign up free for a feed built around what you actually care about, Dive Deeper research on any story, and the full text of every article.
Create free accountAlready have an account? Sign in