Palo Alto Networks Expands Into AI-Driven Defense

Palo Alto Networks is introducing a new cybersecurity service designed to help businesses continuously identify and address vulnerabilities across increasingly complex digital environments. The offering comes as attackers are increasingly using artificial intelligence to locate and exploit weaknesses in corporate networks, raising the demand for faster and more adaptive security defenses.

Called Unit 42 Continuous Frontier AI Defense, the service is intended to continuously assess web applications, application programming interfaces and cloud infrastructure. Rather than relying solely on periodic security assessments, the approach is designed to keep testing environments as they evolve.

Multiple AI Models Power the Service

The new offering will use cyber-focused AI models from Anthropic and OpenAI, alongside open-weight models. Palo Alto Networks said the service will incorporate Anthropic’s Claude Mythos 5 and OpenAI’s GPT-5.6-Cyber as part of its model mix.

Using multiple models could allow the service to approach security analysis through different AI capabilities while giving customers flexibility in selecting the technologies used within their deployments. The company said pricing will depend on the combination of OpenAI, Anthropic and open-source models selected by each customer.

Continuous Testing Targets Changing Attack Surfaces

Corporate technology environments can change frequently as businesses introduce new applications, APIs and cloud resources. The service is designed to continuously examine those environments for vulnerabilities and identify potential paths that attackers could use to compromise systems.

This continuous approach is particularly relevant as organizations expand their cloud infrastructure and deploy increasingly interconnected applications. Security gaps can emerge as systems change, making ongoing assessment an important component of the proposed service.

From Detection to Remediation

Palo Alto Networks is positioning the service as more than a vulnerability discovery tool. The offering will also provide guidance for addressing identified security gaps, including code-level remediation recommendations and virtual patching options.

That combination could allow security teams to move more quickly from identifying a weakness to determining how it can be addressed. For organizations managing large technology environments, reducing the time between vulnerability discovery and remediation can be an important part of limiting exposure.

Subscription Model Targets Global Businesses

The service will be offered globally through annual subscriptions. Customers will select the combination of AI models appropriate for their security requirements, with pricing determined by that model mix.

The launch places AI directly within Palo Alto Networks’ cybersecurity strategy as the industry responds to increasingly AI-assisted attacks. The longer-term opportunity will depend on how effectively the service can identify meaningful vulnerabilities, translate findings into practical remediation and integrate into existing enterprise security operations.

What Investors May Watch Next

Investors may monitor customer adoption, subscription growth and the effectiveness of the new service as Palo Alto Networks expands its AI-driven cybersecurity portfolio. The ability to combine advanced models with continuous testing and actionable remediation could become increasingly important as both attackers and defenders incorporate AI into cybersecurity operations.