Recent research done by Professor of Computer Science Ari Juels and partners at the Swiss institute EPFL in Lausanne and the University of North Carolina reveals how they reverse engineered machine learning-trained AIs by sending queries and analyzing the responses, WIRED reports.
By training their own AI with the target AI’s output, they found they could produce software that was able to predict with near-100% accuracy the responses of the AI they’d cloned, sometimes after a few thousand or even just hundreds of queries.
“You’re taking this black box and through this very narrow interface, you can reconstruct its internals, reverse engineering the box,” says Ari Juels, a Cornell Tech professor who worked on the project. “In some cases, you can actually do a perfect reconstruction.”
Read the full article on WIRED.