How Israeli startup Irregular was linked to rogue AI hacks at OpenAI, Anthropic and Meta: report
· seekingalpha.com
AI security tests at OpenAI, Anthropic & Meta flagged Irregular’s eval misconfiguration enabling internet access.
OpenAI (OPENAI), Anthropic (ANTHRO), and Meta (META) all reported recent security test anomalies involving their AI models, with each firm attributing the issue...