We have developed multiple AIs that identify threats. These cover both detection in images or in data streams.
New methods have been used to detect anomalies. The emergence of auto-encoders to learn what “normal” looks like and then raise alerts when “abnormal” patterns are detected are transforming the capability.
This approach overcomes the issues with traditional methods requiring examples of issues to learn. We can now see ways to keep pace with emerging threats as well as those we have seen before.