Cercetătorii au folosit Claude pentru a sparge sistemele OpenAI, profitând de două vulnerabilități critice

A security research company successfully breached OpenAI's internal systems using Anthropic's Claude chatbot, exploiting critical vulnerabilities. The Hacktron AI team accessed multiple ChatGPT accounts, reporting their findings to OpenAI, which rewarded them $6,500. They initially used Claude but later relied on OpenAI's GPT-5.6 Sol model. The breach involved exploiting a vulnerability in Discourse, leading to control over OpenAI employees' accounts. This incident highlights ongoing security challenges for OpenAI, which has faced multiple security issues recently. Anthropic has called for a slowdown in AI development, a sentiment echoed by other tech leaders, while Donald Trump criticized such calls as detrimental to U.S. competitiveness in AI.