It’s tedious — and expensive — for humans to scrutinize data looking for abnormalities, and that’s why MIT’s Computer Science and Artificial Intelligence Laboratory developed its AI2 method. AI2 uses artificial intelligence to screen data daily from tens of millions of log lines and then flags anything deemed suspicious. A data analyst checks any of the flags and the method helps identify around 85-percent of attacks and saves time analysts normally spend chasing false leads.
Much like any AI system, AI2 must be trained. When it reports the errors, an analyst lets it know which ones are false meaning its becoming more successful. Most of what AI2 does is reactive, meaning it helps a company determine what happened and how to proceed. However, the learning capabilities AI2 possesses strengthen its intelligence and saves workers time every single day.
Learn More: web.mit.edu