Team breaches OpenAI in 3 days

A team of cybersecurity researchers from "Hactron AI" discovered a vulnerability in the systems of "OpenAI," the developer of the ChatGPT tool, and managed to breach it for less than $3,000 over the course of three days. The team exploited the company's social media platform and a flaw in the "Discourse" platform to generate malicious code, which allowed them to access the company's tools and products and modify its repositories on GitHub. They carried out the breach using "Claude Oops 4.8" and "Oops 5.0" models from "Anthropic." After notifying the company about the vulnerability, which was subsequently patched, they received a monetary reward of $6,500. This breach comes amid growing global concerns about the cybersecurity risks posed by artificial intelligence models, emphasizing the need to strengthen security measures and training to address increasing security challenges.

articleDetails.fontSize.sizeحجم النص

تنويه: هذا ملخص تم إنشاؤه بواسطة الذكاء الاصطناعي

عرض المقال بالكامل