Zeroport

Zeroport's core product is Fantom, a hardware-first remote-access architecture built around a physical separation at the network boundary. The company describes a directional interaction model in which keyboard and mouse events move into the protected environment and compressed screen pixels move out, while file transfer, clipboard, print, and ordinary IP tunneling are absent from the access path. Fantom Edge is positioned for smaller or distributed sites and OT clusters; Fantom Enterprise is a rack-mounted, centrally managed deployment for larger environments. The product therefore combines an isolation mechanism with identity, policy, session monitoring, recording, audit logs, and integrations with SSO, MFA, and security monitoring workflows. These technical details are primarily company-reported and should be verified through architecture review, penetration testing, and customer references.

The customer problem is specific and consequential: operators, administrators, vendors, and incident responders still need interactive access to systems that were designed to be segmented or offline. VPNs, remote desktop gateways, PAM jump hosts, ZTNA brokers, and browser-isolation products can reduce exposure, but they generally preserve an IP-connected pathway and depend on software configuration, patching, and credential controls. Zeroport is selling a different control point for organizations that would rather remove network reachability and data-transfer channels than manage them indefinitely. Its own industry material targets financial services, utilities, industrial and manufacturing environments, data centers, semiconductor facilities, public-sector organizations, and critical infrastructure. The company also says Fantom can be deployed without changes to protected applications or internal routing; that operational claim is important to validate in pilots because deployment, latency, availability, and support requirements will influence adoption more than the architectural story alone.

Commercially, Zeroport appears to be an early but unusually well-capitalized startup. Its official announcement states that it emerged from stealth with a $10 million seed round in January 2026 led by lool ventures, with Clarim Ventures, CyberFuture, and Fusion Fund participating. The company website presents products, industry-specific material, a demo-led sales motion, and international operating presence; its LinkedIn profile lists 11–50 employees, offices in New York, Israel, and Singapore, and a Herzliya headquarters. Those are useful commercialization signals, but they do not establish recurring revenue, deployment scale, retention, independent security certification, or named customer traction. The current diligence question is whether the company can convert high-intent pilots in regulated and industrial environments into repeatable appliance-plus-subscription deployments while maintaining hardware supply, field support, and acceptable total cost of ownership.

The competitive set spans more than direct non-IP substitutes. Cloudflare, Zscaler, Netskope, Okta, BeyondTrust, CyberArk, Microsoft, Palo Alto Networks, Fortinet, and remote-operations vendors compete for portions of the secure-access budget through cloud ZTNA, PAM, VPN replacement, or OT remote-access products. Data diodes and unidirectional gateways are adjacent substitutes in high-segmentation environments, although they typically solve controlled data transfer rather than interactive remote access. Zeroport's strongest differentiation is a physical enforcement point and a constrained human-interface channel; its weakest point is that hardware introduces procurement, redundancy, lifecycle, and integration burdens that cloud-first competitors avoid. Claims such as eliminating attack paths or making classes of attacks impossible must be treated as scoped product claims, not as a blanket security guarantee: the management plane, endpoints, identity provider, appliance firmware, privileged users, and protected hosts remain part of the system's attack surface.

The defense and national-security relevance is credible but not yet independently demonstrated in the public record. Hardware-enforced access to separated networks maps naturally to military facilities, defense contractors, intelligence environments, utility control systems, and government operations where lateral movement or data exfiltration would have outsized consequences. Former military-intelligence experience described on the company site may help with problem selection and procurement context, but it is not evidence of a defense contract or operational deployment. For strategic diligence, the most important next evidence would be independent validation of the hardware boundary, secure-boot and firmware-update practices, cryptographic and management-plane design, fail-safe behavior, measured latency and availability, red-team results, customer references, and proof that the restricted input/output model supports real workflows without creating unsafe workarounds.