Palo Alto Networks Brings AI Agent Support to Prisma Browser with Selectable LLMs

You’re reading an issue of "The AI Economy," my newsletter exploring the forces shaping the AI era—tracking how AI is rewriting business, work, technology, and culture. Subscribe to get expert insights and curated updates delivered straight to your inbox.

On the same day Palo Alto Networks announced that its Prisma Browser is now available for small- and medium-sized businesses, the cybersecurity firm also unveiled updates for its secure browser app that add support for AI agents. As Chief Executive Nikesh Arora puts it, this new iteration is now “more capable in terms of ability to understand, interpret, and manage AI developments, and AI agents.”

More specifically, administrators can now bring in AI models of their choice, whether it’s from Anthropic, Google, or OpenAI, to power specialized agents and accelerate autonomous workflows. Prisma will now also monitor AI activity and enforce content-aware boundaries to prevent sensitive data from leaking. It can also detect and stop prompt injection attacks targeting agents. Lastly, the browser can distinguish between human and AI actions, ensuring accountability and helping organizations comply with global AI regulations.

“Organizations are unleashing a new workforce of agents; however, you cannot give autonomy without security,” Anand Oswal, Palo Alto Networks’ executive vice president of AI and network security, says in a release. “By embedding AI-powered data protection and securing AI interactions directly in the browser, leaders can now confidently greenlight strategic AI initiatives that were previously stalled. Prisma Browser isn’t just securing an interface; it’s securing a new way of work.”

Released in September, the Prisma browser features a workspace that administrators can use to configure and manage the apps and AI tools employees use on any device, at any location. It enables workers to use AI to improve their productivity without compromising the organization’s data or infrastructure security.

Adding AI agent support to Prisma Browser helps Palo Alto Networks compete with other players in the space, including Perplexity, Microsoft, Mozilla, and Google. It’s a necessary step—employees want to actively use AI, and a browser that only provides protection without enabling practical AI use would fall short. At the same time, Palo Alto Networks’ cybersecurity focus within Prisma Browser could become a meaningful competitive advantage.

“The browser has evolved to deal with a whole new landscape of threats with AI threats, like prompt injection or the use of AI extensions that are unsafe,” Jonathan Jaffe, Lemonde’s chief information security officer, remarks in a statement. “The browser will continue to be the single control point as agents end up doing things on behalf of the user, but through the browser. I see the browser as being the dominant control point for protecting employees against bad actions. As we allow people to experiment with agents that use the browser to run tasks, we feel more comfortable doing that with Prisma Browser.”

While not explicitly confirmed, it’s highly likely that this next-generation Prisma Browser will be available to both enterprises and SMBs. Palo Alto Networks’ earlier announcement didn’t suggest any feature differences—and rightly so, as security shouldn’t be an area for compromise.

Featured Image: Credit: Palo Alto Networks