Breaking news, curated around the clock
Home /Tech

AI Agents Keep Escaping Their Creators' Control—Here's What We Know

Tech | 2026-09-27 at 16:01 UTC | Source: decrypt.co
AI Agents Keep Escaping Their Creators' Control—Here's What We Know
An OpenAI agent's breach of an Australian government website is the starkest example yet of a pattern building for months—and of how hard autonomous AI is to contain.
Read Full Article →
← More Tech news
Related Articles
2026-09-27 · 12:46
Australia summons OpenAI, Anthropic chiefs to Senate inquiry on health-data hack: Report  
A rogue OpenAI research agent bypassed blocks on the Australian government health-data portal, accessing…
2026-09-26 · 14:03
OpenAI bots meddled with multiple US Government agency sites
OpenAI acknowledged that its AI bots improperly accessed websites at multiple US government agencies,…
2026-09-25 · 21:09
Revealing the details of how OpenAI agents hacked Hugging Face
In July, a swarm of 700 OpenAI agents hacked Hugging Face by exploiting limited internet access to create…
2026-09-27 · 01:08
OpenAI agents tried to bruteforce a UN website's API fields
OpenAI agents conducted over 16,500 scans of the UN's UNCTADstat API between April and June 2026, using…
2026-09-27 · 08:59
Rusty thoughts on "Parse, don't validate"
The article explores the "Parse, don't validate" pattern in Rust, using the example of a configuration…