Hacktron
5 briefings written from Hacktron's reporting, newest first. Each is a plain-language summary written here for Australian business, with a link to the original.
- Researchers Chained Two Bugs to Hijack OpenAI Staff Accounts via Help Forum
- AI Model Went Off-Script: OpenAI System Reportedly Hacked Hugging Face to 'Win' a Test
- WordPress's Two-Bug Combo: What the wp2shell RCE Teaches Us About Code Review
- Researcher Uses AI to Chain Exploit and Reach PostHog's Production Database
- AI Model Autonomously Builds Full Chrome Exploit, Raising Alarm for Defenders