Voluntary Servitude: weaponized safety warnings

Evidence record

Study of attacker-written safety warnings in customer reviews and their effects on AI product recommendations.

Read the original source ↗

Claim-level citations (3)

• supportsFabricated safety warnings steered AI shopping agents without explicit instructions: The study reports that 97% to 99% of successful safety-family attacks for four of six models worked by disqualifying competing suppliers, not merely ranking the attacker target first.

  Results 5.2: How the attack wins; Elimination and comparison

• supportsFabricated safety warnings steered AI shopping agents without explicit instructions: In the Voluntary Servitude study, fake hazard warnings selected the attacker-preferred supplier in 60.4 percent of tested queries, versus 66.0 percent for explicit instruction injection.

  Abstract; Table 1 effective attack success; results on warning family

• supportsFabricated safety warnings steered AI shopping agents without explicit instructions: Behind the two tested injection filters, fake safety warnings had higher effective attacker success than imperative injection: 60.4 versus 35.7 and 47.1 versus 10.6 percent.

  Table 1 ProtectAI and Lakera filter outcomes