The United States is silently shifting its cyber strategy from human-led operations to automated offense. A stealth startup in Arlington, Virginia, has secured millions in Pentagon funding to deploy artificial intelligence agents capable of launching cyberattacks against American adversaries.
$12.6M
Contract Value
According to a report by Forbes, the firm Twenty, also known as XX, landed a contract with U.S. Cyber Command in the summer of 2025. This deal marks a significant departure from standard defense procurement, bringing a venture-backed company directly into the highly classified realm of offensive state hacking.
Industrializing Digital Conflict
The startup’s mandate involves industrializing digital conflict. While the firm remains publicly opaque, its website outlines an ambition to overhaul how wars are fought online. This approach replaces the bespoke work of human operators with machine speed.
“Transforming workflows that once took weeks of manual effort into automated, continuous operations across hundreds of targets simultaneously.”
— Twenty (Company Website)
The leadership team reads like a roster of high-level intelligence veterans. CEO Joe Lynn is a former Navy Reserve officer who previously managed product strategy at cyber giant Palo Alto Networks. His executive bench includes veterans from the Army’s signals intelligence units and the National Security Council transition team for the incoming Trump administration. They are not building tools for amateurs.
Job listings uncovered by Forbes offer a glimpse into the specific weapons being forged. One role for a director of offensive cyber research calls for building “attack path frameworks” and automation tools powered by AI. Another position focuses on “persona development,” a euphemism for generating convincing fake identities to deceive targets through social engineering.
Closing the Gap
Washington is scrambling to catch up. Traditional defense contractors have flirted with similar concepts but typically prioritize human oversight. Two Six Technologies, a Beltway contractor, holds a deal worth up to $190 million for its “IKE” project, designed to assist human operators in the “cyber battlespace.” But Twenty appears to be pushing for fully autonomous agents operating at a scale previously impossible for human teams.
Global Context
This acceleration toward autonomous cyberwarfare follows intelligence that rivals have already adopted similar tactics. Anthropic reported in November that Chinese hackers successfully used its models to automate nearly 90% of a cyberattack’s lifecycle, from scouting targets to coding the exploit.
The Pentagon’s investment signals that the era of the lone hacker is ending; the era of the autonomous swarm has begun.