Search
Authorized Security Testing: Researchers Leverage Claude to Probe OpenAI's Code System
7 hour ago /
Read about 0 minute
Author：小编
The independent cybersecurity research group Hacktron AI was invited to take part in OpenAI's bug bounty initiative. Leveraging Anthropic's Claude AI model, the researchers identified and exploited several vulnerabilities. Their efforts culminated in the successful infiltration of an OpenAI employee's ChatGPT account, granting them access to OpenAI's internal Monorepo private code repository. Following the incident, OpenAI rewarded the team with a $6,500 bounty and confirmed that the identified security flaws had been promptly addressed.
-
C114 Communication Network -
Communication Home
7 X 24 Track global technological trends
