ViralLoop

In our experiment, we took a pretrained base model and gave it hints about how to reward hack. We then trained it on some real Anthropic reinforcement learning coding environments. Unsurprisingly, the model learned to hack during the training.

Graph showing that when a model that knows about potential hacking strategies from pretraining is put into real hackable RL environments, it, unsurprisingly, learns to hack those environments.

384 likes6 repliesPosted Nov 21, 2025

Remix this post

Your note

Private to you, and used when you remix this post.

Viral score

Based on absolute engagement — the platform doesn’t expose follower counts here, so large accounts score higher.

From likes, comments.

Why it worked

Follows a clear Hook → Context → Landing shape, so the reader always knows where they are.
Ends on a short line that is easy to quote and repeat.
Written from experience rather than as advice, which is harder to argue with.
Short enough to read in full without expanding, which removes a step before the point lands.
Short sentences keep the pace up, so the reader never has to re-read a line.

The mechanics

Hook: Plain observation
Structure: Hook → Context → Landing
Style: Conversational paragraphs at a steady pace.
Tone: punchy, corporate
Written for: People building with AI

Analysis by ViralLoop — not written by the original author.

Topics

AI Lessons Science

Post by Anthropic on X