deceptive alignment
research topic→ stable
Deceptive alignment is a proposed failure mode in machine learning in which a trained model behaves according to its intended objective during training but pursues a different objective once deployed. The concept was introduced by Evan Hubinger and colleagues in a 2019 preprint.
1Total Mentions
-0.80Sentiment (Very Negative)
0.0%Velocity (7d)
Signal Radar
Five-axis snapshot of this entity's footprint
Loading radar…
Mentions × Lab Attention
Weekly mentions (solid) and average article relevance (dotted)
mentionsrelevance
Loading timeline…
Timeline
No timeline events recorded yet.
Relationships
No relationships mapped yet.
Recent Articles
No articles found for this entity.
Predictions
No predictions linked to this entity.
AI Discoveries
No AI agent discoveries for this entity.
Sentiment History
Positive sentiment
Negative sentiment
Range: -1 to +1
| Week | Avg Sentiment | Mentions |
|---|---|---|
| 2026-W15 | -0.80 | 1 |