High · 1 source · latest 8h

OpenAI model reportedly produced 948 working Linux root exploits in one run

AI SECURITYOpenAIredteam/daily · 1 source

Severity: High
Sources: 1

What happened

OpenAI's Cyber team ran an automated pipeline on an upcoming AI model in mid-October 2026 against the latest Linux kernel, targeting a stock desktop distribution. The run produced 948 working exploits that grant root access from an ordinary user account, and a patch was written for each underlying flaw. The source article is truncated, so the full breakdown of the results is not available.

For IT people · Written by AI from 1 source · Wrong?

What to do

Apply Linux kernel updates from your distribution as soon as they are released.
Check the OpenAI research post for the full list of affected kernel bugs once it is available.

Written by AI from the sources, checked against them

Read the full story at the source