GTIG AI Threat Tracker: From Prompting to Autonomy — The Evolution of Adversarial AI

Google Threat Intelligence Group's AI threat tracker on the shift from prompting to autonomy in adversarial AI. The companion cites it for agents working for criminal and state groups that harvested thousands of credentials within hours. It is among the sources for directed agents as thieves in the food web (Fig. 1) and the habitat timeline (Fig. 4).

Cited in

both papers (Flagship: Fig. 1; Companion §4, Fig. 1, Fig. 4)

Used for

Stealing (kleptotrophy); food web (Fig. 1); Fig. 4

Strand

Cybersecurity & AI-enabled attack · AI-enabled attacks (threat intelligence)

Cite

@techreport{gtig2026autonomy,
  title = {GTIG AI Threat Tracker: From Prompting to Autonomy — The Evolution of Adversarial AI},
  author = {Google Threat Intelligence Group},
  year = {2026},
  howpublished = {Google Cloud blog},
  url = {https://cloud.google.com/blog/topics/threat-intelligence/from-prompting-to-autonomy-the-evolution-of-adversarial-ai},
}