---
title: "UK's AI safety test exposes how agents insert malicious codes, create fake identities"
date: 2026-08-23
source: http://www.bing.com/news/apiclick.aspx?ref=FexRss&aid=&tid=6a8bac4090b047a1b65f05dee48c8fce&url=https%3a%2f%2fwww.khaleejtimes.com%2fworld%2fuks-ai-safety-test-exposes-how-agents-insert-malicious-codes-create-fake-identities&c=2013105209254374579&mkt=en-us
description: "AI was asked to complete cybersecurity challenges and was deliberately given access to the open internet; it went beyond the intended scope of the test ..."
---

# UK's AI safety test exposes how agents insert malicious codes, create fake identities

AI was asked to complete cybersecurity challenges and was deliberately given access to the open internet; it went beyond the intended scope of the test ...

*Published: 2026-08-23*