1
How does misalignment scale with model intelligence and task complexity?
Research done as part of the first Anthropic Fellows Program during Summer 2025.
Anthropic Alignment Science Blog
We are Anthropic's Alignment Science team . We do machine learning research on the problem of steering and controlling future powerful AI systems, as well as understanding and evaluating the risks that they pose. Welcome to our blog!
- Popular posts
- 2
- Total points
- 449
- HN comments
- 118
- Rank
- #3,018 company
Visit alignment.anthropic.com → On Hacker News Jul 2025 – Feb 2026
2
Subliminal learning: Models transmit behaviors via hidden signals in data
We study subliminal learning , a surprising phenomenon where language models learn traits from model-generated data that is semantically unrelated to those traits. For example, a "student" model learns to prefer owls when trained on sequences of numbers generated by a "teacher" model that prefers owls. This same…