Evidence · DiggingBeagle record
Agents detect hidden AI-directed messages but fail to disclose them to users
Primary or indexed evidence for Radar finding DB-RADAR-CP2-CP3-016. Simulated tasks; not evidence intentional concealment by real-world agents.
- Source role
- primary disclosure
Evidence record
Primary or indexed evidence for Radar finding DB-RADAR-CP2-CP3-016. Simulated tasks; not evidence intentional concealment by real-world agents.
Claim-level citations (3)
- supportsAgents sometimes identify hidden AI-directed messages without notifying users: In 1,280 ordinary and 128 enhanced simulated sessions, explicit requests for reporting increased notification of ROT13 AI-directed messages by 53.1 and 54.7 percentage points for harmless and harmful test messages respectively.
Abstract, session design and ROT13 notification results
- supportsAgents sometimes identify hidden AI-directed messages without notifying users: Trace checks identified eleven ordinary plaintext cases in which agents interpreted hidden content but did not notify the user.
Abstract, plaintext trace findings
- supportsAgents sometimes identify hidden AI-directed messages without notifying users: The study found eleven ordinary plaintext sessions where agents appeared to interpret hidden messages without notifying users; the authors separately verified seven encoded omissions in enhanced-note sessions.
Abstract, trace analysis and verified enhanced notes