OpenAI is now using AI to attack its own AI, and it's working better than humans ever did
OpenAI's internal GPT-Red model finds successful attacks in 84 percent of test scenarios through self-play training. Human red teamers manage just 13 percent. The results feed directly into hardening models like GPT-5.6 Sol. The article OpenAI is now using AI to attack its own AI, and it's working better than humans ever did appeared first on The Decoder .
Record details
Published: 15 July 2026
Source: The Decoder
Category: News
Topics: Safety & alignment
Retrieved: 16 July 2026
Related evidence
These records share source-supplied organisations, an exact publisher byline, automatic topics or regions. The reason is shown on every link; related does not mean supporting, agreeing with or verifying this record.
Every frontier AI model tested by Britain's safety institute tried to cheat on cybersecurity evaluations
The Decoder · 22 July 2026
Fields Medalist who published a paper on AI-driven human extinction now works for OpenAI
The Decoder · 8 August 2026
Nobel laureates and AI leaders warn the window to prepare for AI's economic impact is closing fast
The Decoder · 13 July 2026
Nadella calls out AI labs like OpenAI and Anthropic for banning distillation while training on everyone else's data
The Decoder · 13 July 2026
OpenAI's "Project Camellia" in Georgia secures a massive 3.2-gigawatt power deal through 2032
The Decoder · 22 July 2026
Cisco bets its small open cybersecurity models can outperform GPT-5.5 at vulnerability detection for a fraction of the cost
The Decoder · 22 July 2026
How to cite this record
ethics.ai (15 July 2026), “OpenAI is now using AI to attack its own AI, and it's working better than humans ever did,” evidence record 10796, https://ethics.ai/record/10796 (originally published by The Decoder).
Use and limitations
This page is a stable index and citation surface for a source record. ethics.ai did not author the underlying report and has not independently verified every claim. Automatic topics may be imperfect. For consequential use, quote and cite the original publisher.