OpenAI breached by researchers using Anthropic models

Security researchers successfully breached OpenAI by utilizing tools from its rival, Anthropic. The attackers leveraged Anthropic's Claude models to execute the hack into the AI lab. Following the incident, OpenAI provided the researchers with a $6,500 bug bounty. This security breach underscores the escalating cybersecurity threats associated with advanced AI agents, demonstrating how competing AI technologies can be weaponized to compromise the infrastructure of other leading artificial intelligence organizations in the industry.