Anthropic weighs new model before IPO amid antitrust lawsuit over safety collusion claims

A finance news account reported that Anthropic is considering a new model release while OpenAI advances in enterprise. CEO Dario Amodei has called for slower development on safety grounds. Separately, the company posted measurements for public visibility into frontier labs. One account summarized details from the post, including Claude leading a portion of internal R&D work, full monitoring of research agents, and a share of compute devoted to safety studies. The post itself states the goal is to let the world see what happens inside AI labs.