Pangea is a company that developed an AI Guardrail Platform to help security and product teams ship secure AI applications quickly. The platform also helps security teams protect workforce AI use with guardrails that span prompt injection detection, sensitive data leakage detection, AI visibility, AI access controls, and agentic security controls. Pulse 2.0 interviewed Pangea founder and CEO Oliver Friedrichs to learn more about the company.
Oliver Friedrichs’s Background
What is Oliver Friedrichs’ background? Friedrichs said:
“I’ve spent my career building and scaling cybersecurity companies—some names you might recognize, like SecurityFocus, Immunet, and Phantom, which found homes at Cisco, Symantec and Splunk. While it’s more of an art than a science, you get good at spotting blind spots in the security space before they become headlines. My background spans both offensive and defensive security, which helps to see vulnerabilities that are quietly emerging in our AI-fueled future. That’s where I spend most of my time now—trying to keep up with what’s next and then how to protect it.”
Formation Of The Company
How did the idea for the company come together? Friedrichs shared:
“The concept for Pangea originated from identifying a recurring pattern in technological adoption: security consistently lags behind innovation cycles. This pattern manifested in the early internet, cloud computing, mobile technologies, and now AI. The name “Pangea” references the prehistoric supercontinent that existed before tectonic separation—a deliberate metaphor for our mission to unify the fragmented world of cybersecurity.”
Favorite Memory
What has been your favorite memory working for the company so far? Friedrichs reflected:
“Beyond the tech, we’re proud of initiatives like The Great AI Escape—a virtual escape room that challenged the community to solve real-world LLM security problems in a hands-on, gamified format. But above all, my greatest joy is the team we’ve built—passionate, mission-driven people who are shaping the future of AI security.”
Core Products
What are the company’s core products and features? Friedrichs explained:
“Pangea is a security platform purpose-built for modern app developers—especially those working with AI, cloud-native, and mobile-first architectures. Our products accelerate app delivery by providing essential guardrails for building secure AI and cloud applications.”
“One of our biggest milestones has been the launch of our AI security suite, which analyzes user and system prompts in real time to detect and block jailbreak attempts with over 99% accuracy across over 160 prompt injection methods. It’s a powerful, layered defense that’s raising the bar for enterprise AI protection. We’ve also earned early recognition from industry leaders and landed our first major enterprise customers who were grappling with AI-related security gaps.”
Challenges Faced
What challenges have Friedrichs and the team face in building the company? Friedrichs acknowledged:
“AI is evolving faster than any tech cycle we’ve seen before, and that creates a moving target for security. One of the biggest challenges is how fast the threat landscape shifts—especially with adversarial AI. We’ve tackled this by building a culture that prioritizes speed and adaptability, and by creating a platform that can evolve in tandem with the threats. We also stay tightly connected with our partners and the broader research community—it’s not something you can solve in a vacuum.”
Evolution Of The Company’s Technology
How has the company’s technology evolved since launching? Friedrichs noted:
“We’ve come a long way since 2021. Initially, we focused on securing cloud-native applications—but we quickly saw the need to expand into AI security as the use cases and risks exploded. Today, our platform includes targeted detection and mitigation capabilities for generative AI-specific threats, including things like prompt injection. We’re constantly iterating to stay ahead of the curve.”
Customer Success Stories
When asking Friedrichs about customer success stories, he highlighted:
“One of the most compelling examples is how Grand Canyon Education—who supports 22 major U.S. universities—used Pangea to roll out GenAI chatbots securely at scale. Their CISO had a clear goal: enable AI-powered innovation across their student and staff platforms, without risking exposure of sensitive student data to external AI models like OpenAI or Anthropic.”
“Instead of locking down AI use—which would have slowed innovation to a crawl— the team deployed our API-driven redaction solution. It automatically strips out sensitive data from user prompts and uploaded files in real time, before anything ever leaves the organization’s environment. It’s a smart, pragmatic approach that shows what’s possible when security works hand-in-hand with enablement. This isn’t just a win for GCE—it’s a blueprint for how enterprises everywhere can embrace AI responsibly.”
Funding
When asking Friedrichs about the company’s funding details, he revealed:
“Pangea has raised $51 million total over two rounds of funding. I can’t disclose specific revenue figures at this time.”
Total Addressable Market
What total addressable market (TAM) size is the company pursuing? Friedrichs assessed:
“The broader AI infrastructure market is expected to exceed $96 billion by 2027. Within that, we’re laser-focused on securing enterprise AI apps, especially in sectors where the stakes are high. When you add up the risk, the market opportunity, and the regulatory tailwinds, we’re looking at a TAM that’s not just large, but essential.”
Differentiation From The Competition
What differentiates the company from its competition? Friedrichs affirmed:
“Pangea’s key differentiation lies in its breadth – protecting against an industry’s leading 8 of the top 10 OWASP LLM risks. To do this we developed a comprehensive set of security services specifically designed for AI applications, which has enabled us to create sophisticated capabilities like our prompt injection detection system.”
“This system requires continuous updates similar to anti-virus definitions as new attack techniques emerge daily. We’ve also pioneered innovation in additional areas that others haven’t fully addressed, such as our MultiPass library for real-time permission validation during RAG inference, and our format-preserving encryption that allows sensitive information to retain its structure while being protected. This combination of cybersecurity heritage and comprehensive capability development allows us to provide uniquely effective protection as organizations rapidly scale their AI implementations.”
Future Company Goals
What are some of the company’s future goals? Friedrichs concluded:
“Looking ahead, we’re building the industry’s broadest framework to secure AI from cloud to code. This includes agent security—what we see as the next frontier in AI protection. As AI transitions from simple interactions to executing tools and complex chains of thought with planning capabilities (such as Claude 3 Opus or the GPT-4 model), the security implications become far more significant.”
“We’re actively building taxonomies of prompt injection methods. We’ve already classified more than 160 different attack methods, and are constantly retraining our detection models to respond to evolving threats. Long term, we want to help write the rulebook for securing AI—both technically and operationally.”
Additional Thoughts
Any other topics you would like to discuss? Friedrichs concluded:
“One thing I’ll always advocate for is collaboration. AI is moving too fast for anyone to go it alone. We need more shared intelligence, open dialogue, and collective accountability across the public and private sectors. If we get this right, AI can be a force for enormous good. But only if we secure it first.”
