Terra Security, a groundbreaking startup whose proxy AI-powered penetration testing platform reshapes the cybersecurity landscape, has announced a $8 million seed round led by Syn Ventures and FXP Ventures. Additional support comes from underscore venture capital and well-known angel investors, including former Google Ciso Gerhard Eschelbeck and founders of Talon Security Ben-Noon and Ohad Bobrov.
The company has partnered with Fortune 500 clients and plans to use capital to expand its multi-agent capabilities, develop new red team features and accelerate customer adoption.
Turn the table: AI for offensive safety
Defense has historically been prioritized when it comes to cybersecurity, but Terra Security is flipping the script. Its breakthrough comes from leveraging proxy AI (a goal-oriented, semi-autonomous agent) that can simulate large-scale behaviors of skilled hacking. These agents are not universal scripts. They have fine-tuned AI “employees” to constantly explore each customer’s network environment, adapting to business logic, code updates and changes in emerging threats in real time.
The core of the Terra platform is a multi-agent architecture, where dozens of professional AI agents operate in parallel with discovering potential vulnerabilities. Unlike traditional tools that rely on hard-coded inventory, these agents use real-world attack strategies to constantly scan and rescan web applications, such as an opponent who never falls asleep.
To maintain accuracy and reduce false positives, Terra used human models to ensure that AI-generated discoveries are validated and guided by expert human testers. This synergy between machine scalability and human judgment involves one of the biggest drawbacks in traditional pen testing solutions: inconsistent accuracy and lack of context.
Continuous Penetration Testing: A New Gold Standard
Historically, penetration testing has been plot-based, which is an expensive annual event or quarterly compliance checkbox. However, as the enterprise environment develops with dazzling speed, point-of-time assessments leave critical blind spots.
Terra’s continuous penetration test model shifts safety testing from responsiveness to active. Even after discovering smaller changes like new third-party plugins or feature updates, its platform will automatically launch new test scenarios. This is because modern web applications are dynamic, integrating APIs, cloud infrastructure and evolving user flows, which are potential entry points for attackers.
The company’s approach is particularly effective in addressing business logic vulnerabilities, a flaw in the workflow and decision-making process that traditional scanners often miss. By learning the unique background and tailoring test plan for each application accordingly, Terra provides important insights, not just noise.
“The pen test shouldn’t be just a box you check once a year.” explain Shahar depressedCEO and co-founder of Terra Security. “We transform it into a continuity, context and strategic layer of your security posture. Agent AI enables us to simulate real opponents with better coverage and consistency than ever before.”
Why, why now?
The explosion of web-based applications has made organizations more exposed than ever.
This is where Terra stands out. Its agents not only look for the top ten OWASP vulnerabilities, but also identify zero-cycle, API vulnerabilities and multi-step attack chains while adapting to the specific ecosystem of the business. And, unlike traditional tools that cannot rotate like an attacker, Terra’s proxy can link vulnerabilities together, simulate lateral motion, and accurately map the entire attack surface.
Jay Leek, managing partner at SYN Ventures, called Terra “a reimagining penetration test, as we know today, should have been reimagined long ago.”
FXP Ventures, an early believer in the Terra team, responded to this view. “We support Terra from day one because of the founder’s deep technical DNA and ruthless execution.” explain Tsahy Shapsa from FXP. “They not only improve penetration testing — they also redefine it with AI staff working 24/7 at the top human expertise guidance. It’s not humans and machines. It’s Man Plus Machine. It’s the future.”
Used for scales, adjust to accurate
Founded in 2024, Terra Security provides a fully managed platform for offensive security, providing market-leading accuracy, efficiency and web attack surface coverage. Each test plan is tailored to the organization’s risk profile, environment and compliance needs. Whether it is an e-commerce platform facing payment fraud or a fintech application facing the risk of API exploitation, Terra’s AI agents adapt to their surroundings and evolve as the threat changes.
Their platform is particularly important in the following industries:
-
Financial Services – Prevent account acquisitions and ensure complex API workflows.
-
E-commerce – Reduce the risk of payment fraud and compliance failures such as PCI DSS.
-
manufacturing – Protect IOT-based environments from network intrusions.
What’s next for Terra?
Following this round, Terra plans to launch the team feature of proxy red, enabling organizations to perform simulated attacks that go beyond application-level capabilities and mimic complex, full-stack full-stack opponent behavior. It will also expand to network-level testing and wider security assessments to create an all-in-one AI-powered offensive security suite.
Terra Security offers a fascinating new paradigm: Intelligent, lasting AI agents think and behave like hackers and ensure their behavior is accurate, context-sensitive and meaningful with human supervision.
As the cyberweapons race accelerates, Terra offers defenders the first real offensive advantage. With this new capital and ambitious roadmap, the company is in a good position to conduct continuous, intelligent pen testing of the new gold standard for cybersecurity.
