Palo Alto Networks Launches AI-Powered Cybersecurity Service Using Anthropic, OpenAI Models

09:09 AM EDT, 09/22/2026 (MT Newswires) -- Palo Alto Networks (PANW) on Tuesday launched Unit 42 Continuous Frontier AI Defense, an agentic offensive security service designed to continuously identify, validate and remediate enterprise cyber exposures. The service uses a proprietary multi-model AI harness integrating Amazon (AMZN)-backed Anthropic's Claude Mythos 5, Microsoft (MSFT)-backed OpenAI's GPT-5.6-Cyber and open-weight models with Unit 42's threat intelligence and offensive security expertise, it said. The service provides an initial full-estate baseline scan followed by continuous testing as an organization's environment changes, Palo Alto said.