Palo Alto launches frontier AI defence capability

Palo Alto Networks has launched an agentic offensive security service designed to perpetually track down and remediate vulnerabilities in gated AI capability models before attackers can weaponise them.

Unit 42 Continuous Frontier AI Defense is targeted at enterprises seeking to leverage models such as Anthropic’s Claude Mythos and OpenAI’s GPT and who want to limit their exposure. It has been developed to combat the use of AI to find and exploit vulnerabilities by today’s threat actors.

Palo Alto said the use of AI by threat actors is compressing breach cycles from weeks to hours. The solution is designed to help defenders keep pace by offering continuous testing to identify and validate vulnerabilities, pinpoint exposures and attack paths, and then accelerate remediation.

The service includes a continuous testing engine that can provide a full estate baseline scan, followed by always-on testing that adapts to changes in an organisation’s environment. A proprietary multi-mode harness software architecture routes work to the model best suited to the task, and advanced adversary simulation proves real-world exploitability of the vulnerability by validating end-to-end attack paths.

The capability has been developed and validated over six months of in-house testing and across more than 100 Unit 42 customer engagements. Palo Alto says that during the internal testing, the capability found a year’s worth of exposure in just three weeks, while during customer assessments, it found exposures in 100% of customers. Of the exposures uncovered, 37% rated high or critical in severity.

Palo Alto Networks SVP of Unit 42 Sam Rubin said AI has created an asymmetric advantage for threat actors against organisations trying to defend at human speed.

“Modern cybersecurity requires machine-speed defence. We’re combining Unit 42’s offensive testing and threat intelligence expertise with industry-leading AI harnesses and exclusive access to gated capability models to give defenders back the upper hand,” he said.

The capability builds on Unit 42’s Frontier AI Defense, launched in April, which introduced a point-in-time exposure analysis that benchmarks current capabilities to help organisations modernise their cybersecurity systems. Last month, Unit 42 announced it would extend its Frontier AI Exposure Analysis with OpenAI’s GPT-5.6-Cyber and Anthropic’s Claude Mythos 5.

Image credit: iStock.com/mustafaU