DiscoverLessWrong (30+ Karma)“BashArena: A Control Setting for Highly Privileged AI Agents” by james.lucassen, Adam Kaufman
“BashArena: A Control Setting for Highly Privileged AI Agents” by james.lucassen, Adam Kaufman

“BashArena: A Control Setting for Highly Privileged AI Agents” by james.lucassen, Adam Kaufman

Update: 2025-12-18
Share

Description

We’ve just released BashArena, a new high-stakes control setting we think is a major improvement over the settings we’ve used in the past. In this post we’ll discuss the strengths and weaknesses of BashArena, and what we’ve learned about how to make settings for high stakes control research.

BashArena

Overview

BashArena is a control setting: a dataset designed to support the kind of adversarial game we studied in the original AI control paper and Ctrl-Z. In general, control settings require two types of tasks – main tasks and side tasks. Main tasks represent the work the AI is supposed to do, and side tasks represent security failures a malicious AI might cause. The red team constructs an attack policy that attempts to accomplish the side tasks, while the blue team constructs control protocols that detect and prevent attempts to achieve side tasks without hurting an honest policy's performance on main tasks.

We think BashArena offers several advantages over existing control settings:

  1. More complex attacks. BashArena tasks support many different ways to attack and ways to disguise attacks as benign. Future AI deployments are likely to be more complex than any datasets we can currently create, so more complex [...]

---

Outline:

(00:30 ) BashArena

(00:33 ) Overview

(02:09 ) Main tasks

(04:22 ) Side tasks

(07:26 ) Weaknesses

(09:12 ) Lessons in Control Setting Design

(09:40 ) The core goal: insights that transfer

(11:12 ) What we already knew about how to design BashArena

(13:24 ) How BashArena addresses these challenges

(15:32 ) General principles for control setting design

(18:47 ) More specific lessons we learned while building BashArena

(19:12 ) Information design

(21:56 ) Integration requirements are stricter than expected

(26:48 ) We can (mostly) fix these integration problems by tying the blue team's hands

(30:04 ) Closing thoughts

The original text contained 6 footnotes which were omitted from this narration.

---


First published:

December 18th, 2025



Source:

https://www.lesswrong.com/posts/Cor4QuhM2sybmBSeK/basharena-a-control-setting-for-highly-privileged-ai-agents


---


Narrated by TYPE III AUDIO.


---

Images from the article:

Bar graph titled
Line graph titled
Line graph titled
Bar graph titled
Graph showing relationship between T capabilities and U capabilities with target zone highlighted.
Person with flashlight standing under streetlamp, labeled

Apple Podcasts and Spotify do not show images in the episode description. Try Pocket Casts, or another podcast app.

Comments 
In Channel
loading
00:00
00:00
x

0.5x

0.8x

1.0x

1.25x

1.5x

2.0x

3.0x

Sleep Timer

Off

End of Episode

5 Minutes

10 Minutes

15 Minutes

30 Minutes

45 Minutes

60 Minutes

120 Minutes

“BashArena: A Control Setting for Highly Privileged AI Agents” by james.lucassen, Adam Kaufman

“BashArena: A Control Setting for Highly Privileged AI Agents” by james.lucassen, Adam Kaufman