[KongchangAI]
ConceptReasoning Hijacking

Reasoning Hijacking

An attack in which the attacker does not directly command the agent but instead contaminates the reasoning process so that the agent adopts the attacker's objective at an early stage and subsequent actions faithfully execute the contaminated plan.

Timeline (last 90 days)

Oct 19

Reasoning hijacking refers to an attacker tampering with intermediate goals or premises in the agent's reasoning process so that the model 'believes' it is executing a reasonable plan; the final action itself may be fully compliant, with the anomaly appearing only early in the reasoning trace.

Unverified50%

All Facts (1)

Source Articles