MIT Develops AI That Detects 85 Percent of Cyber-Attacks

Researchers from the Massachusetts Institute of Technology have created an AI system that can predict a cyberattack before it happens in 85% of incidents. Analyst-driven systems rely on rules created by people and consequently can't detect attacks that don't adhere to those rules, whereas machine-learning systems rely on anomaly detection, which tends to generate false positives that have to be investigated by people.MIT researchers have announced that they've concocted a new artificial intelligence system capable of successfully detecting 85% of cyber-attacks. Part of the challenge of merging human- and computer-based threat detection has been the manual labeling of data for algorithms.The system has been tested on 3.6 billion log lines or pieces of data that reveal major system activities triggered by millions of users over a period of three months. It then reports this activity to a human analyst who can then judge if there's an actual attack.With that feedback, it takes on board whether or not it should be classifying the events as attacks or not, then refines its internal models.According to Engadget, Kaylan Veermachaneni, co-creator of the system, said that one should think of the new system as a virtual analyst. In the near future the industry and federal regulators will need to figure out a balance between the need of cyber security and protecting consumers' privacy. This method often leads to false positives, meaning that humans doubt the reliability of the system and are forced to go back and check all the results anyway.And the more data it analyses, the more accurate it becomes.