OpenAI releases GPT-6 Cyber with restricted access for approved offensive security work
- Severity
- Info
- Sources
- 1
What happened
OpenAI's GPT-6 Cyber reportedly scored 48.5 percent on ExploitGym against 32.1 percent for GPT-5.6 Cyber, and was rated Critical under OpenAI's Preparedness Framework. OpenAI says it avoided out-of-scope exploitation over 99 percent of the time, with 0.07 percent of cases exceeding scope; details of those cases were not disclosed. The model reportedly found 81 high-severity Firefox vulnerabilities, which Mozilla has started fixing, and helped develop exploits for 948 Linux kernel vulnerabilities, 588 of them previously unknown, with patches submitted for maintainer review. No CVE identifiers were given. Access is restricted to approved Daybreak Red customers.
For security analysts · Written by AI from 1 source · Wrong?
What to do
- No action needed.
Written by AI from the sources, checked against them