hackingincidents.ai
Incidents289Sources240Dataset2026-11-17Last sweep2026-11-17
Ledger / Oct 2024

Hijacking AI infrastructure with non-human identities

Confirmed

Permiso found the activity in its own investigations.

Model
AWS BedrockAmazon · closed
Agent harness
Nonechat or tool use only
Attributed country
Not attributedthe source names no country
Sectors hit
ai platforms, enterprise broadVictim AWS Bedrock organizations

Permiso found attackers hijacking victims' AWS Bedrock infrastructure to power unfiltered roleplay chatbot services, passing the cost to the victims. The date is approximate.

“Permiso has found that some attackers are using hijacked LLM infrastructure to power unfiltered sexual roleplaying AI chatbot services.”Permiso

Sources

Date
Oct 2024
Actor
Unnamed LLMjacking attackers (criminal), as attributed by Permiso
Targets
Victim AWS Bedrock organizations
What the AI did
LLMjacking
Model as named
AWS Bedrock
Model basis
A researcher names it
Autonomy
Assisted a human
Outcome
Victims' Bedrock infrastructure powered attackers' chatbot services at the victims' cost.