OpenAI's latest AI model has new protective measures to prevent biological risks

OpenAI's Latest AI Models Have New Protective Measures to Prevent Biological Risks

via cnBeta.COM Chinese Industry Information Station - Telegram Channel

Telegraph
OpenAI's Latest AI Models Have New Protective Measures to Prevent Biological Risks

OpenAI's latest AI models have new protective measures to prevent biological risks. OpenAI said that the performance of O3 and o4-mini has significantly improved compared to previous models, thus bringing new risks to malicious attackers. According to OpenAI's internal benchmarking, o3 is more proficient in answering certain types of biological threat-related questions. For this reason, and to reduce other risks, OpenAI has created a new monitoring system, which the company describes as a "security-centric inference monitor". This monitor runs on o3 and o4-mini, is custom-trained to reason about OpenAI's content policies. It aims to identify content related to biological and chemical risks...