OTT Cybersecurity LLC Launches Significant Innovations for AI Security with Lyrie.ai
In a notable development for the cybersecurity landscape, OTT Cybersecurity LLC, a Dubai-based firm, has announced two transformative milestones in its journey to create foundational infrastructure for the emerging agentic AI era. These milestones include acceptance into Anthropic’s Cyber Verification Program (CVP) and the public unveiling of the Agent Trust Protocol (ATP), a pioneering open cryptographic standard aimed at securing AI agents operating autonomously online.
Leading the charge at OTT Cybersecurity, CEO Guy Sheetrit expressed confidence in their newly accepted status within Anthropic’s CVP, viewing it as validation of the technology that underpins Lyrie.ai. “Being among the first companies accepted into Anthropic’s Cyber Verification Program validates what we’ve built. Lyrie isn’t a security tool that sits alongside AI. It’s the security layer that AI runs on top of,” Sheetrit remarked. This vision underscores Lyrie.ai’s role not merely as an auxiliary security tool but as an essential foundational layer for autonomous AI operations.
The Rise of Autonomous AI Agents and Security Challenges
The rapid deployment of autonomous AI agents by enterprises and government bodies presents a dual-edged sword; while these agents possess the ability to read emails, write code, transfer funds, and sign contracts, this exponential increase in digital autonomy has raised substantial concerns around security. Traditional security models have fallen short in addressing these complexities on an enterprise scale. Recognizing this gap, Lyrie was developed as a comprehensive solution to empower organizations to securely manage and monitor the actions of autonomous AI agents.
Central to Lyrie’s offering is the newly released Agent Trust Protocol (ATP). The ATP, crafted by Lyrie’s research team and now publicly available at lyrie.ai/research, allows systems to verify, in real-time, the identity of AI agents, their authorized actions, and whether any tampering has occurred. This capability is crucial as it addresses the fundamental uncertainty of engaging with autonomous agents that can potentially act in unpredictable ways.
The ATP is centered around five essential primitives:
- Identity — Clearly defining who the AI agent is.
- Scope — Outlining what actions the agent is authorized to undertake.
- Attestation — Assessing whether the agent or its instructions have been altered.
- Delegation — Identifying who has given the authority to the agent.
- Revocation — Confirming whether that authority has been rescinded.
Guy Sheetrit highlights a pressing concern in contemporary digital interactions: “Every AI agent on the internet today is a stranger. You don’t know who it is, what it’s authorized to do, or whether it’s been tampered with. ATP is the protocol that changes that.”
The Agent Trust Protocol is set to revolutionize how organizations interact with AI agents, providing a layer of trust that is currently lacking in the digital ecosystem. Importantly, ATP is an open, royalty-free standard, with its reference implementation available on GitHub under the MIT license, allowing developers to integrate this crucial technology into their systems.
Strategic Partnerships and Future Pathways
OTT Cybersecurity’s acceptance into Anthropic’s Cyber Verification Program opens doors for collaboration within the wider cybersecurity community. The program aims to verify legitimate dual-use cybersecurity operators and will facilitate Lyrie’s ongoing work related to vulnerability research and offensive security tooling within Claude’s AI infrastructure. Subject to Anthropic’s safety and security policies, this partnership promises to enhance Lyrie’s capabilities as it seeks further verification pathways with other leading AI laboratories.
About Lyrie and Its Security Innovations
Lyrie.ai serves as a comprehensive offensive and defensive cybersecurity platform designed specifically for the AI-driven landscape. It includes features like:
- Lyrie Hack — An autonomous penetration testing command that follows a structured, seven-stage workflow, generating proof-of-concept exploits and remediation strategies.
- GPU-Powered Red Teaming — A suite of adversarial testing workflows leveraging advanced GPU infrastructure.
- OWASP ASI 2026 Alignment — Ensuring threat coverage aligns with the OWASP Agentic Security Initiative taxonomy.
- Omega-Suite Binary Analysis — Focused on discovering zero-day vulnerabilities in compiled software.
- Flexible Deployment Architecture — Scalable across various hardware systems, from consumer-grade devices to enterprise-level GPU clusters.
- Integrated Security Toolkit — Featuring an array of tools for reconnaissance, exploitation, and remediation, all within a unified agent environment.
OTT Cybersecurity LLC, based in the United Arab Emirates, is committed to developing robust cybersecurity solutions that stem from real-world operational insights. By addressing the unique challenges of the AI era, the company aims to lead the charge in securing a future where autonomous systems can operate safely and effectively.
For more information about their offerings and initiatives, interested parties are encouraged to visit Lyrie.ai and explore their research at this link. For inquiries, Guy Sheetrit can be reached via email at guy@lyrie.ai.