Timezone: »

Deceptive Alignment Monitoring
Andres Carranza · Dhruv Pai · Rylan Schaeffer · Arnuv Tandon · Sanmi Koyejo
Event URL: https://openreview.net/forum?id=obsO44GFhh »

As the capabilities of large machine learning models continue to grow, and as the autonomy afforded to such models continues to expand, the spectre of a new adversary looms: the models themselves. The threat that a model might behave in a seemingly reasonable manner, while secretly and subtly modifying its behavior for ulterior reasons is often referred to as deceptive alignment in the AI Safety & Alignment communities. Consequently, we call this new direction Deceptive Alignment Monitoring. In this work, we identify emerging directions in diverse machine learning subfields that we believe will become increasingly important and intertwined in the near future for deceptive alignment monitoring, and we argue that advances in these fields present both long-term challenges and new research opportunities. We conclude by advocating for greater involvement by the adversarial machine learning community in these emerging directions.

Author Information

Andres Carranza (Stanford University)
Andres Carranza

Hi! I'm Andres: a Colombian student studying at Stanford University interning at Two Sigma and previously at NASA.

Dhruv Pai (Computer Science Department, Stanford University)
Rylan Schaeffer (Stanford University)
Arnuv Tandon (Computer Science Department, Stanford University)
Sanmi Koyejo (Stanford University)

Related Events (a corresponding poster, oral, or spotlight)

More from the Same Authors