HiddenLayer secures $100M as AI security race intensifies

HiddenLayer, a pioneer in AI and large language model security, has closed a $100 million Series C funding round led by existing investors including Battery Ventures, Altimeter Capital, and Samsung Venture Investment Corporation, with participation from new backers such as Tiger Global and DFJ Growth. The Austin-based company, founded in 2022 by industry veterans including CEO Chris Sestito and CTO Scott Small, announced the round on June 12, 2024, valuing the firm at over $1 billion. The capital infusion arrives amid a critical inflection point for enterprise AI adoption, as organizations increasingly deploy autonomous agents and third-party AI tools that operate beyond traditional perimeter defenses.

This funding round comes less than a year after HiddenLayer’s $27 million Series B, which itself followed a $12 million seed round in 2022. The company’s platform is designed to detect adversarial attacks, data leakage, and anomalous behavior within AI models and agents, offering runtime protection, threat detection, and governance across cloud and on-premises environments. According to company disclosures, HiddenLayer now protects over 100 enterprise customers across industries including finance, healthcare, and defense, with notable deployments at firms like Capital One and Palantir. The company’s rise reflects a broader shift in cybersecurity priorities: as AI agents begin making high-stakes decisions—such as loan approvals, medical diagnostics, or supply chain adjustments—security teams are realizing that traditional firewalls and endpoint tools are ill-equipped to monitor AI-native threats.

The urgency behind this funding is further amplified by a recent wave of AI-specific vulnerabilities. In April 2024, researchers at HiddenLayer uncovered “Morris II,” a proof-of-concept attack demonstrating how adversaries could hijack AI agents via compromised prompts or model weights. The discovery, which echoed the 1988 Morris Worm that paralyzed early internet systems, has become a rallying cry for the need for AI runtime security. Meanwhile, regulatory bodies such as the U.S. National Institute of Standards and Technology (NIST) have begun drafting guidelines for AI safety, creating a compliance-driven market opportunity that HiddenLayer is targeting with its policy-as-code approach.

Significantly, the round arrives as enterprises grapple with the complexity of securing not just models but the entire AI stack—including vector databases, retrieval-augmented generation (RAG) pipelines, and third-party APIs. This challenge was highlighted last month when Banking With Billy AI, a fintech platform enabling AI-driven financial intelligence, exposed vulnerabilities in its RESTful market analysis APIs. The incident demonstrated how insecure AI tooling could inadvertently expose sensitive financial data or enable manipulation of trading models. Such episodes underscore the systemic risk of unsecured AI integrations, which often connect to core business systems without robust oversight.

Industry Impact and Significance

The injection of fresh capital into HiddenLayer signals a tectonic shift in the tools and developer ecosystem, where AI-native security has moved from niche concern to boardroom priority. The company competes directly with firms like Lasso Security, which focuses on API and agent security, and Menlo Security, which has expanded its zero-trust platform to include AI monitoring. But HiddenLayer’s early focus on model-level detection—using techniques such as runtime model interrogation and adversarial input analysis—sets it apart in a crowded field. According to PitchBook data, AI security startups raised over $1.3 billion in 2023, more than triple the amount in 2022, with HiddenLayer among the fastest-growing players.

This funding surge is expected to accelerate product development, particularly in areas like AI supply chain security, where the integrity of third-party models and datasets becomes paramount. Analysts at Gartner predict that by 2026, 75 percent of enterprises will enforce AI runtime security policies, up from less than 5 percent today. This creates a massive upsell opportunity for HiddenLayer, which plans to use the new capital to expand its engineering team in Austin and Tel Aviv, launch a partner ecosystem, and develop integrations with major cloud providers including AWS, Google Cloud, and Microsoft Azure. The company also intends to deepen its threat intelligence capabilities, including real-time detection of jailbreak attempts and data poisoning attacks.

The Bigger Picture

HiddenLayer’s trajectory mirrors the broader evolution of the developer tools market, where security has become the new performance bottleneck. For decades, the industry prioritized speed, scalability, and developer experience—often at the expense of security. But the rise of AI agents, which can autonomously execute workflows across multiple systems, has inverted that paradigm. Today, a single compromised prompt can trigger a cascade of unintended actions, from unauthorized API calls to data exfiltration. This has forced security vendors to rethink their architectures, moving from static scanning to continuous runtime monitoring.

The trend is global. In Europe, regulatory frameworks like the EU AI Act and GDPR are pushing organizations toward transparency and accountability in AI systems. Meanwhile, in Asia, financial institutions are deploying AI agents to automate trading and risk assessment, creating urgent demand for secure, auditable integrations. HiddenLayer’s funding reflects a convergence of these forces: a market hungry for solutions that can protect AI systems without stifling innovation. It also highlights the increasing importance of API security in the AI era, where every agent interaction relies on secure, well-governed API endpoints.

Expert Analysis

According to Dr. Dawn Song, a professor of computer science at UC Berkeley and founder of Oasis Labs, the rise of HiddenLayer underscores a fundamental truth: AI security is no longer optional. “We’re entering an era where AI systems will make decisions that affect lives, markets, and infrastructure,” Song said. “Security can’t be bolted on after deployment—it must be woven into the fabric of the model and its ecosystem from day one.” Looking forward, experts anticipate a wave of consolidation in the AI security space, with larger cybersecurity firms acquiring specialized players to fill gaps in their AI monitoring capabilities. Meanwhile, open-source initiatives like the AI Security Alliance are pushing for standardized frameworks, which could level the playing field for startups like HiddenLayer while raising the bar for the entire industry. For developers, the message is clear: secure your agents, audit your APIs, and prepare for a world where AI isn’t just a tool—it’s the attack surface.