Palo alto networks launches ai-driven cybersecurity service to counter AI threats

Palo Alto Networks has launched a new subscription service that leverages a mix of frontier and open-weight AI models to identify vulnerabilities in customers' networks. The service, set to roll out on Tuesday, includes models like Anthropic's Mythos 5 and OpenAI's GPT-5.6-Cyber. The company aims to detect security flaws and recommend fixes before potential cyberattacks. In internal testing, no single AI model caught more than 40% of vulnerabilities in complex environments. Arora, a representative from Palo Alto Networks, emphasized that combining multiple AI models with human expertise is the future of cybersecurity. As AI adoption expands across enterprises, the company plans to monitor where AI is concentrated to better secure these areas. The move highlights the ongoing arms race between AI-powered attackers and defenders in the cybersecurity landscape. The initiative underscores the growing reliance on AI for both offensive and defensive cyber operations. With AI models becoming increasingly sophisticated, security vendors are under pressure to innovate and respond at scale. Palo Alto Networks' approach reflects a broader industry trend of integrating AI into cybersecurity strategies. The company's focus on enterprise AI adoption signals a shift toward more targeted and proactive security measures. As AI continues to transform various sectors, the need for robust defenses will only intensify, driving further investment and innovation in cybersecurity solutions.