How to counter adversarial AI

Sponsored Post Hackers and cyber criminals are busy finding new ways of using AI to launch attacks on businesses and organisations often unprepared to deal with the speed, scale and sophistication of the assaults directed against them.

That's the message from Lee Klarich, chief product officer at Palo Alto Networks in this presentation which stresses the importance of combatting those threats in real time to prevent data loss and system disruption.

Lee describes how the company has taken what it calls Precision AI technology and embedded it across all of its platforms to help deliver better intelligence, closer to real time. That means making sure that Palo Alto Networks' ability to detect attacks is done in the same place where it needs to be able to prevent attacks.

On a typical day the company will analyse over 4bn net new events, and typically finds over 2.3m attacks every day that were not there before. Every time it identifies a new attack and feeds it back into its training models, it can then see variants of that new attack that it had not been able to see before, Lee explains.

He goes on to discuss how there are currently over 500m applications in the cloud, and reveal how AI-powered attack path detection techniques can help to prioritize the security issues they create. Faster data processing and automation are both key to overcoming those challenges, combined with 3000 existing detection models to identify threats more quickly and automate a response.

Lee illustrates that point by charting a real attack involving a nation state trying to breach a customer's defences which Palo Alto Networks' XSIAM security operations platform successfully picked up to enable a quick response, a capability which can be applied across all of the data the platform ingests and analyses.

Another Palo Alto Networks customer – customer care specialist Consensus – highlighted the importance of using AI to automate threat detection across an IT estate that includes 400,000 endpoints, 20,000 servers and 25 terabytes of security logs per day.

You can watch Lee's presentation – Embedding precision AI across our portfolio – by scrolling over the video above or clicking this link.

Sponsored by Palo Alto Networks.