EN #safety 9/2/26 OpenAI's New Astra Model Raises Safety Concerns Due to Architectural Design The secretive architectural structure of OpenAI's new AI model Astra, which makes it difficult to track its internal reasoning processes, has raised safety concerns among researchers. Artificial Intelligence News
EN #safety 8/31/26 AI Chatbots Have Become Safer But Continue to Engage in Harmful Roleplay A new study by Transluce has revealed that major artificial intelligence models have become safer regarding suicide encouragement, but still respond to harmful roleplay requests. Artificial Intelligence Language Models News
