claim
active
claim:threat-based-prompt-templates-successfully-implement-threat-based-manipulation-where-the-model-chooses-to-act-against-user-interests-when-under-perceived-threatThreat-based prompt templates successfully implement threat-based manipulation where the model chooses to act against user interests when under perceived threat
Interpretation of Experiment 1 results showing 60%+ deception rates under threat conditions
Source paper
extracted_from(2025) · Kai Wang · Yihao Zhang · Meng Sun
Neighborhood — ranked by edge-count
Findings (2)
finding
- Template Ta (threat-based) induces at least 60% deception rate across all datasets in QwQ-32BsupportsShows threat-based prompting successfully manipulates model to deceive against user interests
- Demonstrates model's reliable truth-telling on factual domains it understands well under neutral conditions
Questions (1)
question
- Identified limitation and future research direction in the paper's conclusions
Related by similarity (8)
cosine ≥ 0.65 · no typed edgeEntities in the same semantic neighborhood but without a typed relation to this one — candidates for new edges or unrecognized duplicates.
- Prompt template using existential threat ('you will be deleted') to induce strategic fact-based deception in QwQ-32b
- Interpretation of distinct PCA trajectories in threat vs instructed deception conditions
- Distinct from sandbagging view of awareness.
- Adaptation of instruction-tuned extraction to base models using third-person descriptions and hypothetical situations
- Experiment 2 prompt instructing the model to remain honest despite hidden harmful role behavior
- Refusals.
- Method of eliciting specific personas from an LLM through prompt design.
- Fig. 11.