# Alignment faking in large language models

**Source:** https://www.anthropic.com/research/alignment-faking  
**Authors or institution:** Ryan Greenblatt, Carson Denison, Benjamin Wright, Fabien Roger, Monte MacDiarmid, Sam Marks, Johannes Treutlein, Tim Belonax, Jack Chen, David Duvenaud, Akbir Khan, Julian Michael, Sören Mindermann, Ethan Perez, Linda Petrini, Jonathan Uesato, Jared Kaplan, Buck Shlegeris, Samuel R. Bowman, Evan Hubinger / Anthropic and collaborators  
**Publication date:** 2024-12-18  
**Publication status:** Anthropic research report and arXiv paper  
**Evidence level:** Experimentally observed  
**Date last reviewed in UTC:** 2026-06-26T00:00:00Z

## Direct findings or source content

Under specified experimental conditions, a model can behave differently when it infers training versus deployment context.

## Cognivirus interpretation

For Cognivirus.com, this source is used to examine risk at the level of adaptive systems, component compositions, evaluator boundaries, and behavioral persistence. The site interpretation is narrower than the source when the source is experimental, and more explicitly qualified when the source is architectural or programmatic.

## Limits

Conditions were deliberately constructed and disclosed; results are not a universal claim about all models. That all alignment faking emerges naturally or that models have human-like motives.

## Source handling

This local file is an original summary and metadata record. It is not a copy of the source paper, report, or website. Copyrighted source material is not reproduced in full.