Modelyo raises $17 million to protect data during cloud and AI processing
The AMW Read
The funding adds a confidential cloud entrant to the infrastructure player map, but the reported scope and lack of adoption or performance metrics support only sub-segment significance.
Modelyo raises $17 million to protect data during cloud and AI processing
Israeli cybersecurity startup Modelyo has raised approximately $17 million in a round led by U.S.-based OlivTree, with participation from Seth Merrin, David Phelps and other investors. Founded in 2024, the company employs 30 people and previously relied on founder funding. The proceeds will support research and development of its Sovereign Confidential Cloud platform and international commercial expansion.
Modelyo sits in the AI infrastructure layer, addressing control over data while workloads execute on third-party systems. Its platform combines confidential computing, hardware-based encryption, continuous environment verification and zero-trust architecture. According to the company, customers can run services without exposing data contents or encryption keys to cloud providers and can verify protections cryptographically. That proposition targets a procurement constraint for governments, banks, defense organizations and healthcare companies: using cloud and AI services while retaining control over sensitive information. The announcement establishes a funded entrant in confidential infrastructure; it does not establish adoption or independently verified security performance.
For builders and investors, the concrete diligence question is whether those protections extend across the services an AI deployment actually uses. Modelyo plans to expand encrypted managed services, improve support for AI systems and autonomous agents, and extend GPU compatibility. Those stated plans make service coverage and deployment complexity useful evaluation criteria. The article provides no customer counts, revenue figures or performance measurements to assess commercial traction or operating tradeoffs.
