Palo Alto Networks Advanced URL Filtering now operates a series of inline cloud-based deep
learning detectors that evaluate suspicious webpage contents in real-time to protect
users against zero-day threats. This includes cloaked websites, multi-step attacks,
CAPTCHA challenges, and previously unseen one-time-use URLs. When the firewall
processes a URL request containing suspicious webpage contents, it forwards the HTTP
response data to the cloud and analyzes the contents of the webpage that are deemed
suspicious and is categorized accordingly. The deep learning detectors and analyzers
used to categorize websites are updated and deployed automatically as Palo Alto
Networks threat researchers improve the detection logic, and does not require the
administrator to download and deploy update packages. Cloud inline categorization is
enabled and configured through the URL Filtering profile and requires an active
Advanced URL Filtering license.