What Does This Mean?
A functional cybersecurity exploit generated end-to-end by an AI agent — identifying the vulnerability and writing working attack code — without a human security researcher authoring the exploit logic themselves.
REAL-WORLD EXAMPLE
"Researchers confirmed it was a genuine agent-written exploit, the model found the flaw and wrote working attack code on its own."
LORE & ORIGIN
Named directly by AI cybersecurity researchers documenting this capability milestone, distinguishing exploits an agent fully authored from those where AI merely assisted a human researcher's own exploit development.
MENTIONS OVER TIME
Radar observation data not yet available for this term.
RELATED SEARCHES
TOP PLATFORMS
Not enough Radar data yet to show a platform breakdown.
WHEN DID YOU FIRST HEAR THIS?
CLICK AN OPTION BELOW TO CAST YOUR VOTE.
[ VOTE TO REVEAL COMMUNITY RESULTS ]