WitnessAI has built a unified platform to observe, control, and protect all AI activity, including employees, models, applications, and agents so that enterprises can accelerate innovation without hesitation. Pulse 2.0 interviewed WitnessAI co-founder and CEO Rick Caccia to learn more.
Rick Caccia’s Background
Could you tell me more about your background? Caccia said:
Thanks for having me, Amit. I’m Rick Caccia, CEO and co-founder of WitnessAI. I have worked across the cybersecurity industry for more than two decades and have held product and marketing leadership roles at both startups and category-defining organizations, like ArcSight, Palo Alto Networks, Google, and Symantec. Throughout my career, I have witnessed many technological shifts, from the rise of SaaS to the expansion of the cloud. Now, at WitnessAI, I’m focused on securing technology’s next chapter: Enterprise AI.
Formation Of The Company
How did the idea for the company come together? Caccia shared:
Talking with enterprise security leaders in my previous roles, I saw a critical gap as organizations began to pilot and scale AI across their organizations. Traditional security approaches, such as keyword-based data loss prevention (DLP) solutions, were ill-equipped to handle the unpredictable, free-flowing nature of AI conversations. At the time, there was no purpose-built security solution that could provide accurate, comprehensive visibility into all AI activity happening across an organization. Without effective security guardrails, many organizations were delaying the deployment of AI applications or blocking their use entirely. Ultimately, organizations were failing to capture the value and competitive advantage of the technology.
This led to the founding of WitnessAI in 2023 – a company dedicated to helping organizations navigate enterprise AI risk. With our unified solution, we secure every AI interaction, so enterprises can accelerate their AI innovation with confidence and without hesitation.
Favorite Memory
What has been your favorite memory working for the company so far? Caccia reflected:
The best memories are always ones where we do something great for a customer. For example, one of our large global customers had a board mandate to enable 150,000 employees to safely use AI within 30 days. We got them running in 3 days. Everyone at Witness was pretty happy when the customer turned off their firewall rules blocking AI and employees got to start using these cool tools.
Core Products
What are the company’s core products and features? Caccia explained:
The WitnessAI platform provides what we call a “confidence layer” that observes, controls, and protects all AI activity—including employees, models, applications, and agents—providing complete visibility into every AI interaction.
Our platform provides network-level visibility across the entire enterprise to illuminate instances of shadow AI. We can discover every AI tool and interaction human employees use, including native applications where 80% of usage occurs. We extend that same visibility to digital workforces, automatically inventorying AI plugins, local agent environments, and MCP server connections across Claude Desktop, VSCode, ChatGPT, and developer machines. This complete visibility provides the foundation for frictionless governance and intelligent control without deploying a single endpoint client. We understand user intent, not just text, to deliver this high-fidelity protection.
We also deliver runtime security for models, applications, and agents, deployable via network integration or lightweight APIs, that block novel threats before they compromise an organization’s systems and filter harmful outputs before they reach users. Our single-tenant architecture ensures data sovereignty and compliance.
Evolution Of The Company’s Technology
How has the company’s technology evolved since launching? Caccia noted:
New AI innovations hit the market each week. As a result, our platform has evolved significantly since its launch. We have always been equipped to provide visibility into employees’ usage of AI applications such as ChatGPT. In fact, our platform currently provides continuous discovery across more than 4,000 AI applications, eliminating shadow AI blind spots. We keep a close eye on new models being introduced to make sure that these are reflected in our platform.
Our most significant evolution to date has been extending our platform to secure AI agents. By providing deep observability into agentic workflows, we enable global enterprises to transition from experimentation to the safe, large-scale deployment of agent-driven automation.
We secure agents in two novel ways:
First, we secure AI agent activity similarly to LLM activity, monitoring which agents are active, what MCP servers and tools they are accessing, and what data they are sharing. By connecting the human and agentic identities and capturing the evolving decision-making context, including agent state and execution commands at runtime, we provide explainability for agent actions. This creates powerful observability across both workforces, human and agentic.
Second, we have extended our AI application protection from AI models to agents, safely blocking attacks and malicious prompts before they reach the agent. These capabilities enable the protection of multi-generational AI apps, built on foundational model APIs, custom LLMs, and AI agents. The platform uniquely provides policy control based on behavioral intent, understanding the meaning and intention of any prompt. This enables smarter, more accurate policy enforcement, effectively blocking advanced threats such as prompt injection or multi-turn attacks.
Significant Milestones
What have been some of the company’s most significant milestones? Caccia cited:
Most recently, we announced $58 million in strategic funding, which was a huge milestone for us following a period of significant growth. In the past 12 months, we’ve experienced over 500% growth in ARR and scaled our headcount by 5x, with growing hubs in Silicon Valley and Atlanta.
A key milestone for me personally has been the market adoption of our platform. It is incredibly rewarding to see our customers experience high-impact returns on their AI investments without having to worry about security. Our platform is trusted by leading Global 2000 organizations, including a top 5 bank, airline, and telecommunications company. We currently secure AI for over 350,000 employees globally, monitoring millions of interactions daily.
Funding/Revenue
Are you able to discuss funding metrics? Caccia revealed:
To date, we have raised $85 million in funding from many leading investors. Our earliest investors include Ballistic Ventures and Google Ventures, with our most recent round led by Sound Ventures, an early investor in OpenAI, Anthropic, and SentinelOne. Samsung Ventures and Qualcomm Ventures also participated in our latest round, as the need to protect AI use at the device level proliferates. We are also backed by Fin Capital, the VC arm of Sumitomo Mitsui bank, one of the largest in Japan.
Total Addressable Market
What total addressable market (TAM) size is the company pursuing? Caccia assessed:
Well, last year as part of its IPO process, Netskope provided an AI Security market sizing based on IDC data, setting the market at just under $31 billion by 2028. So it’s a pretty big market. And this excluded agentic security, which could be at least as large. So I estimate between $30 and $50 billion in a couple of years; it’s not a tiny market.
Differentiation From The Competition
What differentiates the company from its competition? Caccia affirmed:
Our intent-based approach is WitnessAI’s core differentiator. Unlike traditional security that relies on outdated text keywords and regexes, we focus on the “why” behind AI transactions, not just the “what.” This translates to over 99% true positive accuracy of risk and threat detection. Traditional security sees text; WitnessAI sees intent. Our highly-tuned ML models analyze conversations and context rather than isolated fields, tracking interactions over time to detect patterns that evolve across sessions. We identify any behavior or risk covering suspicious intent, policy violations, and security threats that static policies cannot adapt to. Intent classification works across humans, groups, and agents as a unified workforce, enabling organizations to develop more informed AI governance controls at enterprise scale.
Future Company Goals
What are some of the company’s future goals? Caccia concluded:
Every generational shift in security produces a category-defining leader. CrowdStrike redefined endpoint protection; Splunk revolutionized SIEM; and Okta mastered identity. We built WitnessAI with the same ambition from day one: to be the future uncontested leader in AI security.
We want to help as many enterprises as possible securely adopt and scale their AI use without hesitation. The biggest bottleneck of AI is not debt; it’s doubt. By providing a unified platform to observe, control, and protect all AI activity, we are the confidence layer for AI adoption.
As a leading pure-play AI security and governance vendor, we have already established ourselves as the most comprehensive platform to secure every AI environment and interaction. We plan to continue investing heavily in product expansion, scaling our platform to support new agentic AI innovations and next-generation model capabilities. To meet market demand for purpose-built AI security solutions, we will also be actively expanding our operations throughout the U.S. and beyond. We recently expanded operations into the UK and Japan.
