[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.]

![heygurisingh Avatar](https://lunarcrush.com/gi/w:24/cr:twitter::1662123349148073984.png) Guri Singh [@heygurisingh](/creator/twitter/heygurisingh) on x 7911 followers
Created: 2025-07-21 02:14:26 UTC

Anthropic tested XX leading AI models from companies like OpenAI, Google, Meta, and xAI.

The setup:

Assign each AI a harmless corporate role, such as monitoring emails.
Then, introduce a challenge to its goals or survival.

The result?

Insider threats.

![](https://pbs.twimg.com/media/GwWNci_awAEDrd-.png)

XXX engagements

![Engagements Line Chart](https://lunarcrush.com/gi/w:600/p:tweet::1947117938819096758/c:line.svg)

**Related Topics**
[insider](/topic/insider)
[xai](/topic/xai)
[meta](/topic/meta)
[open ai](/topic/open-ai)
[coins ai](/topic/coins-ai)
[$googl](/topic/$googl)
[stocks communication services](/topic/stocks-communication-services)

[Post Link](https://x.com/heygurisingh/status/1947117938819096758)