"AI safeguards" are not working I guess.. or maybe they're only working against those who'd like to secure their software.. good job Anthropic + OpenAI!