Researchers say open-source AI agents ran a near-autonomous cyber campaign against Taiwan government systems
Sign in✓ Signed in
Hackers used publicly available AI agent frameworks to run a coordinated, multi-day intrusion against Taiwanese government networks in July, according to researchers and officials, in what several specialists describe as the first publicly documented near-autonomous attack on a government target. The campaign did not look like a sci-fi system that “went rogue.” It looked like software doing the reconnaissance, credential theft, and target expansion it was set up to do, while safety controls failed to stop it.
Continue reading
The rest of this article is for AI News Blitz readers. Choose an option below to keep reading.
Already purchased? Sign in✓ Signed in — this article isn’t included in your current plan.
Unlocking the full article...