Palo Alto Networks Launches AI Cybersecurity Service Using Claude and GPT to Fight AI Cyberattacks

Palo Alto Networks Deploys Frontier AI Models for Continuous Security Testing

Palo Alto Networks has launched Unit 42 Continuous Frontier AI Defense, an AI-powered cybersecurity service designed to combat the accelerating threat landscape created by attackers wielding advanced AI tools1

1

2

2

. The service integrates tightly controlled AI models including Anthropic Claude Mythos 5 and OpenAI GPT-5.6-Cyber, alongside open-weight models, to identify security vulnerabilities across corporate systems1

1

. This marks a significant shift in how organizations defend against threats that now operate at machine speed rather than human speed.

AI-Driven Threats Demand Machine-Speed Defense

The urgency behind this launch stems from a stark reality: attackers using AI have reduced the time from discovering a security flaw to exploiting it by nearly 97% in some cases, compressing what once took weeks into mere hours2

2

. Sam Rubin, Senior Vice President of Unit 42 at Palo Alto Networks, emphasized that "AI has created an asymmetric advantage for threat actors against organizations trying to defend at human speed. Modern cybersecurity requires machine-speed defense"2

2

4

4

. The service addresses this imbalance by providing continuous testing for vulnerabilities as digital environments evolve, rather than relying on periodic assessments that leave gaps between scans.

Multi-Model Approach Uncovers Hidden Security Flaws

Palo Alto Networks' testing revealed a critical insight: no single AI model caught more than 40% of vulnerabilities in complex enterprise environments3

3

. Even more telling, the vulnerabilities discovered by Claude Mythos 5 and GPT-5.6-Cyber overlapped less than 10% of the time3

3

. This finding underscores the value of the multi-model defense strategy embedded in the service. Unit 42's proprietary software intelligently routes tasks to different models, optimizing both detection capabilities and cost efficiency2

2

. Michael Moore, Cybersecurity Lead at Anthropic, noted that "Claude Mythos found flaws that survived decades of human review, and more than ten thousand high-severity vulnerabilities across the software the world runs on"2

2

.

Comprehensive Testing Across Enterprise Attack Surfaces

The Unit 42 Continuous Frontier AI Defense service continuously tests web applications, application programming interfaces, cloud infrastructure, source code repositories, and network assets1

1

4

4

. After conducting an initial baseline scan of a customer's entire estate, the system maintains vigilance as environments change, attempting to validate that each weakness can actually be exploited and mapping potential attack paths an intruder could follow2

2

3

3

. The service then provides actionable guidance, including code-level fixes and virtual patching options to address security gaps before official patches become available1

1

3

3

.

Proven Results Through Extensive Development and Testing

Palo Alto Networks invested $17 million and six months developing this approach across more than 100 customer engagements2

2

4

4

. When deployed on its own systems, three weeks of testing surfaced as many exposures as the company would typically expect to find in an entire year2

2

. Customer assessments conducted with the earlier point-in-time version revealed exposures at every organization tested, with 37% of those vulnerabilities rated high or critical severity2

2

4

4

. This builds on Unit 42's Frontier AI Defense offering introduced in April, which began incorporating these cyber-focused AI models in August4

4

.

Flexible Pricing and Global Availability for Enterprise Customers

The service is available globally through annual subscriptions, with pricing structured around the mix of Anthropic, OpenAI, and open-source models customers select1

1

2

2

. All subscription tiers include access to the multi-model harness system that orchestrates task routing across different AI models4

4

. Palo Alto Networks serves 75,000 customers worldwide, giving the service immediate reach across a substantial portion of the enterprise security market4

4

.

Future Implications for AI Security and Enterprise Defense

CEO Nikesh Arora highlighted that the rapid evolution of AI capabilities outpaces security development, stating that "the capability and technology continues to evolve much faster than you can create security for it"3

3

. He emphasized that combining multiple AI models with human expertise represents the industry's future direction, particularly since each model trains on different data sets3

3

. Looking ahead, Arora indicated that Palo Alto Networks will monitor where enterprise AI adoption concentrates to determine what needs securing next, noting that "if you believe that half of what's going to happen or be possible with AI hasn't been invented yet, then half of that security is missing, too"3

3

. This positions continuous AI-powered security testing as essential infrastructure for organizations navigating an increasingly AI-augmented threat landscape where defenders must match attackers' speed and sophistication.