OpenAI Unveils GPT-5.4-Cyber for Improving Cyber Defense With AI
2026-04-16T02:52:14Z•d5cb5326333bcd9ad9f596603de87c1d46dff0a175c5665879fc1d3792ef7978
CVE-2026-33032active-exploitationai-securityandroid-trojanav-killingchrome-extensionscve-programgpu-rowhammerinfostealerlaw-enforcement-takedownmalwarenginxphishingransomwaresigned-adwarethreat-actorsvulnerability-managementzero-day-patches
What happened
Infosecurity Magazine roundup (Apr 2026) covering major developments: OpenAI launched GPT-5.4-Cyber to support cyber-defence workflows while Anthropic advances Project Glasswing/Claude-assisted vulnerability discovery. ENISA seeks top-level CVE authority status and CISA advocates greater AI-company participation in vulnerability disclosure. Multiple high-impact incidents and patched/exploited flaws were reported — notably an actively exploited nginx-ui MCP authentication-bypass (CVE-2026-33032, CVSS 9.8), Microsoft patched two zero-days, Fortinet issued an emergency EMS patch, and Ninja Forms'
Why it matters
A reviewed impact interpretation has not been published for this record.
Evidence and limitations
- Source ID
- infosecurity_magazine
- Record identifier
- d5cb5326333bcd9ad9f596603de87c1d46dff0a175c5665879fc1d3792ef7978
- Enrichment time
- 2026-04-16T02:52:14Z
- AI-assisted enrichment
- Yes
This record may overlap with other records. Its enrichment can be incomplete or wrong, and machine assistance was used. Validate consequential decisions against the linked source and your own environment.