Anthropic has delayed its expected initial public offering from October to November as it contends with stronger competition from OpenAI, cheaper open-weight models and a widening debate over AI safety. The company was privately valued at roughly $965 billion after a May funding round and is seeking a public-market valuation of as much as $2 trillion, potentially raising up to $100 billion. Its annualized revenue run rate exceeded $65 billion by the end of July, while it has told investors to expect roughly $190 billion to $200 billion in revenue in 2028. However, OpenAI's GPT-6 Astra has gained enterprise traction, and Chinese models from DeepSeek and Moonshot, along with Meta Platforms' Llama, are intensifying price pressure. Anthropic's revenue and enterprise customer base have expanded rapidly, but computing commitments are also rising, including a SpaceX agreement requiring about $1.25 billion in monthly payments over several years. The company is considering a new model before the IPO, even as Anthropic CEO Dario Amodei calls for slower AI development. OpenAI CEO Sam Altman says OpenAI will not go public in 2026 and has instead raised more than $120 billion at a valuation above $1.2 trillion. Anthropic's listing is expected to test whether public investors will support frontier-AI valuations despite uncertain margins, high capital needs and concerns about the sustainability of rapid model development.