Artemis Security Raises $70M Series A to Build AI-Native Cyber Defense Against Automated Attacks

Cybersecurity right now feels like a game where the clock changed but half the players are still running the old play. The offense figured out acceleration. The defense is still checking the replay. Traditional detection tools were built for a world where humans moved fast. Now the machines are running point, and they do not wait for approval. They do not blink. They do not file tickets. That gap between signal and response is where damage lives, and Artemis decided that gap was unacceptable.

So Shachar Hirshberg, CEO, and Dan Shiebler, CTO, got to work. No nostalgia, no patchwork upgrades. Just a clean decision to build something that moves at the same speed as the threat itself. Six to seven months later, Artemis walks out of stealth not asking for attention, but demanding it, backed by a $70M combined Seed and Series A round led by Felicis, with First Round Capital and Brightmind Partners leaning in again, and firms like Theory Ventures, Lockstep, and Two Sigma Ventures placing their bets where the future is already headed .

There is something telling about who else showed up to this table. Founders of Abnormal AI and Demisto. Former leadership from Splunk. Senior operators from CrowdStrike, Palo Alto Networks, Microsoft, and Okta. People who have already seen what happens when the old model breaks, now investing in what comes after. That is not a casual endorsement. That is pattern recognition with capital behind it.

Artemis is not trying to out-alert the problem. It is modeling entire environments, letting AI hunt, investigate, and respond without waiting for a human to catch up. Adaptive detections instead of static rules. Agentic systems instead of dashboards collecting dust. It is less about seeing the fire and more about making sure it never spreads.

And the market is already answering. Customers like Mercury, Wix, Lemonade, and even Abnormal AI are not experimenting. They are deploying. Quietly, deals are closing in the seven-figure range while most of the industry is still debating what AI security even means.

There is a lesson buried in all of this, and it is not subtle. The companies pulling capital right now are not just building better tools. They are building for a different tempo entirely. When the speed of attack changes, the only real move is to match it or get left explaining why logs were reviewed after the fact.