HiddenLayer lands $100M to secure the AI that companies are rushing into production

AUSTIN — HiddenLayer, an Austin company that builds security tools designed specifically to protect AI models, agents and workflows, has raised a $100 million Series B, one of the larger cybersecurity rounds to close in early September.

Delta-v Capital led the financing, with participation from Ten Eleven Ventures, M12, Booz Allen Ventures and Morgan Stanley. The round brings HiddenLayer's total funding to about $156 million, according to Crunchbase data.

The company said it will use the capital to deepen its enterprise platform, including runtime security for agentic systems and protections around the harnesses that connect agents to other tools, while expanding its channel partnerships and moving further into international markets.

HiddenLayer occupies a category that barely existed a few years ago: defending the AI systems themselves rather than the networks around them. That means guarding against tampering with models, prompt-based attacks, data leakage and the new failure modes that appear when software can take actions on its own.

Why it matters

The urgency is a function of adoption. As companies move generative and agentic AI out of pilots and into production, the blast radius of a compromised system grows from a bad answer to a real-world action, an unauthorized transaction, a leaked record, an agent that goes off-script. Security budgets are following that shift, and a nine-figure round is a marker of how quickly “AI security” has become its own line item.

For Austin, the deal reinforces the city's standing as a hub for security and dual-use software, a category that has drawn outsized investor attention this year. It also lands the round with an investor list that mixes traditional venture firms with strategic backers like Booz Allen Ventures and Morgan Stanley, a sign that enterprise and government buyers are watching the space as closely as the funds are.