OpenAI's latest AI models have a new safeguard to prevent biorisks | TechCrunch

Ahead of o3 and o4-mini, OpenAI says it created a new system to block prompts related to risky biological and chemical subject matter.