Autonomous threat response just got a massive capital injection. Onyx Security, a cybersecurity startup building AI agents designed to detect and remediate enterprise threats without waiting for a human to approve every move, has closed a $113 million funding round, according to Globes reporting. For a company still in its early stages, that is a striking number — and a clear signal of where institutional money thinks the security market is heading.
The bet is on agentic AI: systems that don’t just flag a problem but act on it. That concept is already reshaping adjacent corners of the industry, as seen when Skyhawk Security plugged its own AI red-team engine directly into AWS to hunt threats autonomously. Onyx is pushing the same logic further into enterprise remediation, aiming to compress the window between detection and response from hours or days down to seconds.
What Onyx Is Actually Building
Onyx Security’s platform is built around AI agents that can investigate security incidents and carry out remediation steps across an enterprise environment autonomously. The core problem the company is targeting is well-documented: security teams are overwhelmed by alert volume, and the manual triage process creates dangerous lag between when an attacker moves and when a defender responds. Onyx’s agents are designed to close that gap by handling the investigative legwork and executing containment actions without requiring a human analyst to shepherd each case.
The approach puts Onyx squarely in competition with a crowded but fast-moving segment of the security market. Larger platforms from CrowdStrike and SentinelOne have moved aggressively into automated response, and a wave of AI-native startups is chasing the same enterprise budget. What distinguishes the agentic-AI fringe of that market is the degree of autonomous decision-making involved — these systems are not just recommending actions, they are taking them, which raises both the upside and the compliance stakes considerably.
Why $113 Million, Why Now
The scale of this raise is hard to ignore. A nine-figure round at the early stage reflects how urgently the enterprise security market is being repriced around AI capabilities. Buyers are no longer just asking whether a vendor has an AI feature — they are asking whether the entire response loop can be automated. That shift in buyer expectations is pulling forward enormous investment, and Onyx’s raise is one of the clearest data points yet on how aggressively VCs are willing to fund the answer.
The timing also reflects a broader recalibration in how cybersecurity tools are evaluated. As AI-assisted attacks become faster and more scalable — a concern that has been amplified by cases like the autonomous agent breach that compromised a second company without any human prompt — the pressure on defenders to operate at machine speed is intensifying. A platform that can autonomously investigate and contain incidents is no longer a luxury pitch; it is increasingly the baseline expectation for enterprises dealing with sophisticated threat actors. Onyx’s $113 million suggests investors believe the company has a credible path to delivering that at scale.
No lead investor or valuation figure was disclosed in the Globes report, and Onyx has not published detailed technical benchmarks publicly. But the fundraise alone repositions the company as one of the better-capitalized pure-play agentic security startups in the market — with enough runway to push hard on product development and enterprise sales before the window in this category gets more crowded.
