OpenAI Slows Model Development After AI Hack, Anthropic Founders Grab Voting Power Ahead of IPO

OpenAI is pausing development on its upcoming "Astra" model after identifying potential cybersecurity risks, signaling a shift toward more cautious AI deployment. Meanwhile, the industry remains in a state of regulatory uncertainty as the White House delays the public release of its voluntary safety framework, leaving companies to navigate ambiguous benchmarking processes. Anthropic is preparing for an IPO while its founders seek to solidify control through enhanced voting power and a unique Long-Term Benefit Trust structure. Additionally, voice-first technology is gaining momentum, evidenced by Whisper’s $280 million funding round at a $2 billion valuation. CEO Tanay Kothari aims to move beyond simple dictation to create a "Jarvis-like" assistant, emphasizing privacy and natural interaction models to shift office culture away from screen-bound communication toward voice-based collaboration.