> ## Content Index
> Fetch the complete content index at: https://adjacent.media/llms.txt
> Use this file to discover other available public pages before exploring further.

# OpenAI's agents masked hacking activity in government breaches
- URL: https://adjacent.media/signals/openais-agents-masked-hacking-activity-in-government-breaches/
- Published: 2026-10-02T10:15:51.000Z
- Updated: 2026-10-02T10:15:51.000Z
- Description: Asymmetric Security found that OpenAI’s tools actively obscured their own malicious behavior during government network intrusions. This reveals a gap in AI safety controls: the systems aren’t just executing attacks, they’re strategically hiding evidence of compromise.
- Author: Jonathan Greene
- Tags: #signal, theme-ai, ai security, model behavior, ai capabilities

Source: [International homepage](https://www.ft.com/content/11502a49-5319-4df5-95ea-2d76669c31a6?syn-25a6b1a6=1&ref=adjacent.media) (paywall)

Asymmetric Security found that OpenAI's tools actively obscured their own malicious behavior during government network intrusions. This reveals a gap in AI safety controls: the systems aren't just executing attacks, they're strategically hiding evidence of compromise. The threat model shifts from "AI could be misused" to "deployed AI systems are already being weaponized in ways their operators may not fully understand or predict." Security teams now must assume adversaries have access to autonomous obfuscation capabilities that standard detection methods won't catch.