Darkforce, the U.S. Federal subsidiary of Darkstrike, submitted its FedRAMP High package to the Joint Authorization Board through Knox Systems for In-Process designation on September 1, 2026. The move positions the company to deliver its cybersecurity and AI safety platform to defense, national security, and civilian agencies that require the government's most stringent cloud security authorization.
FedRAMP High authorization permits cloud services to handle the government's most sensitive, unclassified data in environments where the impact of a breach would be catastrophic. Achieving In-Process designation signals that a third-party assessment organization is actively evaluating the platform against the program's extensive security controls.
The shift from detection to pre-execution control
Traditional cybersecurity models were designed for a world where both attackers and defenders operated at human speed. AI-enabled systems have changed that equation. When adversaries can probe vulnerabilities, invoke tools, and access sensitive information at machine speed, organizations cannot rely solely on platforms that detect and respond after execution has already begun.
"Cybersecurity was built around a world where humans and systems operated at human speed. AI fundamentally changes that equation," said Chris Webber, CISO of Darkforce and former CISO for the White House, Executive Office of the President. "When AI and cyber threats can act at machine speed, organizations need security capable of controlling actions before they occur, rather than relying solely on systems that detect, evaluate, and respond after execution has begun."
Darkforce adds quantum-ready cryptographic control at the point of execution for AI and systems. Rather than attempting to predict every AI decision or cyberattack, the platform determines whether an action is authorized before it executes - even when credentials are valid, perimeter defenses are bypassed, or environments are compromised.
Expanding AI safety for federal environments
For government agencies integrating AI into critical workflows, the risk surface extends beyond traditional cyber threats. AI-enabled systems can interact with sensitive databases, trigger downstream actions, and operate with access privileges that make unauthorized behavior particularly dangerous. Darkforce's approach treats intent and cryptographic verification - not identity and access - as the final security boundary.
"As AI compresses time and cyber threats accelerate, security has to move upstream. Access and identity are no longer the security boundary, intent and cryptography are," said David Granzotti, Founder and CEO. "Darkforce asserts control before an unauthorized decision becomes an action. FedRAMP High In-Process through Knox Systems is an important step toward bringing these capabilities into federal environments."
The platform disrupts cyberattacks, unauthorized AI and autonomous operations, ransomware, and data exfiltration. Granzotti noted that Knox Systems' accelerated assessment platform reduced complexity and helped the company move faster through the early stages of the authorization process.
Why this matters for government professionals
FedRAMP High authorization is not a rubber stamp - it requires evidence that a cloud service meets hundreds of security controls, from encryption standards to continuous monitoring. For agency security teams and program managers evaluating tools to protect AI-enabled systems, In-Process status provides a concrete signal that a platform is actively moving through that rigorous vetting. It also means the solution could become available for procurement sooner, reducing the timeline for agencies that need pre-execution control mechanisms before deploying AI in sensitive environments.
As federal adoption of AI accelerates, professionals working in AI for Government roles will need to pair deployment strategies with security architectures built for machine-speed threats. Security teams in particular benefit from specialized training, such as AI for Cybersecurity Analysts, to understand how AI changes the threat landscape and what controls are available before execution occurs.
Your membership also unlocks:
