[GUEST ACCESS MODE: Data is scrambled or limited to provide examples. Make requests using your API key to unlock full data. Check https://lunarcrush.ai/auth for authentication information.]

![AnthropicAI Avatar](https://lunarcrush.com/gi/w:24/cr:twitter::1353836358901501952.png) Anthropic [@AnthropicAI](/creator/twitter/AnthropicAI) on x 597.2K followers
Created: 2025-07-24 17:21:59 UTC

New Anthropic research: Building and evaluating alignment auditing agents.

We developed three AI agents to autonomously complete alignment auditing tasks.

In testing, our agents successfully uncovered hidden goals, built safety evaluations, and surfaced concerning behaviors.

![](https://pbs.twimg.com/media/GwoxhZiWkAsuePd.jpg)

XXXXXXX engagements

![Engagements Line Chart](https://lunarcrush.com/gi/w:600/p:tweet::1948433493102403876/c:line.svg)

**Related Topics**
[coins ai](/topic/coins-ai)
[coins ai agents](/topic/coins-ai-agents)

[Post Link](https://x.com/AnthropicAI/status/1948433493102403876)