The hardware behind the future


Guardrail Technologies Raises Nearly $3 Million in Seed Funding as AI Agents Gain the Power to Act

Filter Fund Venture Capital leads two seed rounds for the AI security startup, which is expanding into Asia and Europe.

Isometric illustration of a shield with a keyhole above an AI core on a railed platform, representing AI security and control

PARK CITY, Utah — Guardrail Technologies, an AI security and control company, said Tuesday it has closed two seed rounds totaling nearly $3 million. Filter Fund Venture Capital led the rounds, and strategic angel investors with senior backgrounds at major global technology companies also took part.

The company says its Australian operations are now running. It plans to use the money to expand into Asia and Europe and to hire across the business.

Securing AI that acts, not just answers

Guardrail is aimed at a problem that grows as AI systems move from generating text to taking actions. Software agents increasingly hold passwords and permissions, and they have the authority to act inside an organization’s systems. The company says its platform lets customers see what their AI is doing, protect it and stay in control of its actions. It pitches the product to high-accountability industries that want to adopt AI quickly without losing oversight.

“This funding lets us meet the moment we built for,” said T.J. Marlin, Guardrail’s CEO. He said Filter Fund “invested before the market saw the problem.” He credited that early backing with helping the company reach issued patents, government validation and clients on two continents.

A founder from the post-mortem side of security

Filter Fund founder Richard Sussman, who is also Guardrail’s Executive Chairman, pointed to Marlin’s background. Marlin previously headed a Big Four forensic technology practice, where Sussman said he “spent his career leading the post-mortems after the major incidents,” working out what caused them and what went wrong. Sussman said that experience shaped Marlin’s vision for an end-to-end platform covering AI security, governance and control.

Sussman argued that fragmented tools and traditional governance models can’t keep pace with AI. “Enterprises need a control layer that evolves at the speed of AI,” he said, adding that he believes Guardrail has built one.

The investor

Filter Fund says Sussman has worked with founders at more than 235 startups, collectively valued at over $11 billion. Its current fund holds 22 companies across AI legal technology, AI cybersecurity, AI governance and AI medical technology. The firm says the fund is up 60 percent to date. That figure comes from Filter Fund and has not been independently verified.

Why it matters

Security for autonomous AI is becoming its own product category. The risk is no longer only that a model gives a wrong answer. An agent with credentials and permissions can take wrong actions, and organizations need a way to monitor and constrain them. Guardrail is betting that a dedicated control layer, separate from the AI vendors themselves, will be part of how companies deploy autonomous systems.

Details on funding, fund performance and company claims are from Guardrail Technologies’ announcement. More at guardrail.tech and filterfund.vc.

Leave a Reply

Discover more from Autonomy Magazine

Subscribe now to keep reading and get access to the full archive.

Continue reading