Anthropic and OpenAI sound the alarm on AI safety and seek to shape how it’s controlled

Anthropic and OpenAI have raised alarms regarding artificial intelligence safety and are actively seeking to influence how the technology's development and control are managed. This heightened concern coincides with a technological shift as AI companies transition from creating standard chatbots to developing advanced AI "world models" featuring 3D awareness.

Amid these rapid capability advancements, intense debates have emerged across the industry regarding the appropriate methodologies and frameworks for testing these increasingly sophisticated technologies before deployment.