Researchers at the Massachusetts Institute of Technology (MIT) have published a paper on a new artificial intelligence platform called AI2 that uses human input combined with machine learning to reduce false positives and increase its ability to predict cyberattacks.
Developed by MIT’s Computer Science and Artificial Intelligence Laboratory (CSAIL) AI2 can predict cyberattacks with an 85 percent accuracy rate, the paper stated, according to the MIT News. The AI combs through the data looking for suspicious activity and then combing and displaying these in meaningful patterns that are then shown to a human analyst. The human than confirms or denies that what was found was malicious.
The MIT team said not only can the platform be scaled up enabling it to review billions of lines of code per day, but the AI learns and becomes more accurate as the system analyzes each attack it endures.